Change search
Refine search result
12 1 - 50 of 52
CiteExportLink to result list
Permanent link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Rows per page
  • 5
  • 10
  • 20
  • 50
  • 100
  • 250
Sort
  • Standard (Relevance)
  • Author A-Ö
  • Author Ö-A
  • Title A-Ö
  • Title Ö-A
  • Publication type A-Ö
  • Publication type Ö-A
  • Issued (Oldest first)
  • Issued (Newest first)
  • Created (Oldest first)
  • Created (Newest first)
  • Last updated (Oldest first)
  • Last updated (Newest first)
  • Disputation date (earliest first)
  • Disputation date (latest first)
  • Standard (Relevance)
  • Author A-Ö
  • Author Ö-A
  • Title A-Ö
  • Title Ö-A
  • Publication type A-Ö
  • Publication type Ö-A
  • Issued (Oldest first)
  • Issued (Newest first)
  • Created (Oldest first)
  • Created (Newest first)
  • Last updated (Oldest first)
  • Last updated (Newest first)
  • Disputation date (earliest first)
  • Disputation date (latest first)
Select
The maximal number of hits you can export is 250. When you want to export more records please use the Create feeds function.
  • 1.
    Almeida, Diogo
    et al.
    KTH, School of Computer Science and Communication (CSC), Robotics, perception and learning, RPL. KTH.
    Ambrus, Rares
    KTH, School of Computer Science and Communication (CSC), Computer Vision and Active Perception, CVAP.
    Caccamo, Sergio
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Chen, Xi
    KTH.
    Cruciani, Silvia
    Pinto Basto De Carvalho, Joao F
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Haustein, Joshua
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Marzinotto, Alejandro
    KTH, School of Computer Science and Communication (CSC), Computer Vision and Active Perception, CVAP.
    Vina, Francisco
    KTH.
    Karayiannidis, Yannis
    KTH.
    Ögren, Petter
    KTH, School of Engineering Sciences (SCI), Mathematics (Dept.), Optimization and Systems Theory.
    Jensfelt, Patric
    KTH, School of Computer Science and Communication (CSC), Robotics, perception and learning, RPL.
    Kragic, Danica
    KTH, School of Computer Science and Communication (CSC), Robotics, perception and learning, RPL.
    Team KTH’s Picking Solution for the Amazon Picking Challenge 20162017In: Warehouse Picking Automation Workshop 2017: Solutions, Experience, Learnings and Outlook of the Amazon Robotics Challenge, 2017Conference paper (Other (popular science, discussion, etc.))
    Abstract [en]

    In this work we summarize the solution developed by Team KTH for the Amazon Picking Challenge 2016 in Leipzig, Germany. The competition simulated a warehouse automation scenario and it was divided in two tasks: a picking task where a robot picks items from a shelf and places them in a tote and a stowing task which is the inverse task where the robot picks items from a tote and places them in a shelf. We describe our approach to the problem starting from a high level overview of our system and later delving into details of our perception pipeline and our strategy for manipulation and grasping. The solution was implemented using a Baxter robot equipped with additional sensors.

  • 2.
    Almeida, Diogo
    et al.
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Karayiannidis, Yiannis
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL. Dept. of Electrical Eng., Chalmers University of Technology.
    Cooperative Manipulation and Identification of a 2-DOF Articulated Object by a Dual-Arm Robot2018In: / [ed] IEEE, 2018Conference paper (Refereed)
    Abstract [en]

    In this work, we address the dual-arm manipula-tion of a two degrees-of-freedom articulated object that consistsof two rigid links. This can include a linkage constrainedalong two motion directions, or two objects in contact, wherethe contact imposes motion constraints. We formulate theproblem as a cooperative task, which allows the employment ofcoordinated task space frameworks, thus enabling redundancyexploitation by adjusting how the task is shared by the robotarms. In addition, we propose a method that can estimate thejoint location and the direction of the degrees-of-freedom, basedon the contact forces and the motion constraints imposed bythe object. Experimental results demonstrate the performanceof the system in its ability to estimate the two degrees of freedomindependently or simultaneously.

  • 3.
    Almeida, Diogo
    et al.
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL. Royal Inst Technol KTH, Sch Elect Engn & Comp Sci, Robot Percept & Learning Lab, SE-10044 Stockholm, Sweden..
    Karayiannidis, Yiannis
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL. Royal Inst Technol KTH, Sch Elect Engn & Comp Sci, Robot Percept & Learning Lab, SE-10044 Stockholm, Sweden.;Chalmers Univ Technol, Dept Elect Engn, SE-41296 Gothenburg, Sweden..
    Cooperative Manipulation and Identification of a 2-DOF Articulated Object by a Dual-Arm Robot2018In: 2018 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA), IEEE Computer Society, 2018, p. 5445-5451Conference paper (Refereed)
    Abstract [en]

    In this work, we address the dual-arm manipulation of a two degrees-of-freedom articulated object that consists of two rigid links. This can include a linkage constrained along two motion directions, or two objects in contact, where the contact imposes motion constraints. We formulate the problem as a cooperative task, which allows the employment of coordinated task space frameworks, thus enabling redundancy exploitation by adjusting how the task is shared by the robot arms. In addition, we propose a method that can estimate the joint location and the direction of the degrees-of-freedom, based on the contact forces and the motion constraints imposed by the object. Experimental results demonstrate the performance of the system in its ability to estimate the two degrees of freedom independently or simultaneously.

  • 4.
    Almeida, Diogo
    et al.
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Karayiannidis, Yiannis
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL. Chalmers University of Technology.
    Folding Assembly by Means of Dual-Arm Robotic Manipulation2016In: 2016 IEEE International Conference on Robotics and Automation, IEEE conference proceedings, 2016, p. 3987-3993Conference paper (Refereed)
    Abstract [en]

    In this paper, we consider folding assembly as an assembly primitive suitable for dual-arm robotic assembly, that can be integrated in a higher level assembly strategy. The system composed by two pieces in contact is modelled as an articulated object, connected by a prismatic-revolute joint. Different grasping scenarios were considered in order to model the system, and a simple controller based on feedback linearisation is proposed, using force torque measurements to compute the contact point kinematics. The folding assembly controller has been experimentally tested with two sample parts, in order to showcase folding assembly as a viable assembly primitive.

  • 5.
    Björklund, Linnea
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Knock on Wood: Does Material Choice Change the Social Perception of Robots?2018Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    This paper aims to understand whether there is a difference in how socially interactive robots are perceived based on the material they are constructed out of. Two studies to that end were performed; a pilot in a live setting and a main one online. Participants were asked to rate three versions of the same robot design, one built out of wood, one out of plastic, and one covered in fur. This was then used in two studies to ascertain the participants perception of competence, warmth, and discomfort and the differences between the three materials. Statistically significant differences were found between the materials regarding the perception of warmth and discomfort

  • 6.
    Blom, Fredrik
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Unsupervised Feature Extraction of Clothing Using Deep Convolutional Variational Autoencoders2018Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    As online retail continues to grow, large amounts of valuable data, such as transaction and search history, and, specifically for fashion retail, similarly structured images of clothing, is generated. By using unsupervised learning, it is possible to tap into this almost unlimited supply of data. This thesis set out to determine to what extent generative models – in particular, deep convolutional variational autoencoders – can be used to automatically extract representative features from images of clothing in a completely unsupervised manner. In reviewing variations of the autoencoder, both in terms of reconstruction quality and the ability to generate new realistic samples, results suggest that there exists an optimal size of the latent vector in relation to the image data complexity. Furthermore, by weighting the latent loss and generation loss in the loss function, it was possible to disentangle the learned features such that each feature captured a unique defining characteristic of clothing items (here t-shirts and tops).

  • 7.
    Brucker, Manuel
    et al.
    German Aerosp Ctr DLR, Inst Robot & Mechatron, D-82234 Oberpfaffenhofen, Germany..
    Durner, Maximilian
    German Aerosp Ctr DLR, Inst Robot & Mechatron, D-82234 Oberpfaffenhofen, Germany..
    Ambrus, Rares
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL. KTH Royal Inst Technol, Ctr Autonomous Syst, SE-10044 Stockholm, Sweden..
    Marton, Zoltan Csaba
    German Aerosp Ctr DLR, Inst Robot & Mechatron, D-82234 Oberpfaffenhofen, Germany..
    Wendt, Axel
    Robert Bosch, Corp Res, St Joseph, MI USA.;Robert Bosch, Corp Res, Gerlingen, Germany..
    Jensfelt, Patric
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL. KTH Royal Inst Technol, Ctr Autonomous Syst, SE-10044 Stockholm, Sweden..
    Arras, Kai O.
    Robert Bosch, Corp Res, St Joseph, MI USA.;Robert Bosch, Corp Res, Gerlingen, Germany..
    Triebel, Rudolph
    German Aerosp Ctr DLR, Inst Robot & Mechatron, D-82234 Oberpfaffenhofen, Germany.;Tech Univ Munich, Dep Comp Sci, Munich, Germany..
    Semantic Labeling of Indoor Environments from 3D RGB Maps2018In: 2018 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA), IEEE Computer Society, 2018, p. 1871-1878Conference paper (Refereed)
    Abstract [en]

    We present an approach to automatically assign semantic labels to rooms reconstructed from 3D RGB maps of apartments. Evidence for the room types is generated using state-of-the-art deep-learning techniques for scene classification and object detection based on automatically generated virtual RGB views, as well as from a geometric analysis of the map's 3D structure. The evidence is merged in a conditional random field, using statistics mined from different datasets of indoor environments. We evaluate our approach qualitatively and quantitatively and compare it to related methods.

  • 8.
    Buda, Mateusz
    et al.
    Duke Univ, Dept Radiol, Sch Med, Durham, NC 27710 USA.;KTH Royal Inst Technol, Sch Elect Engn & Comp Sci, Stockholm, Sweden..
    Maki, Atsuto
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Mazurowski, Maciej A.
    Duke Univ, Dept Radiol, Sch Med, Durham, NC 27710 USA.;Duke Univ, Dept Elect & Comp Engn, Durham, NC USA..
    A systematic study of the class imbalance problem in convolutional neural networks2018In: Neural Networks, ISSN 0893-6080, E-ISSN 1879-2782, Vol. 106, p. 249-259Article in journal (Refereed)
    Abstract [en]

    In this study, we systematically investigate the impact of class imbalance on classification performance of convolutional neural networks (CNNs) and compare frequently used methods to address the issue. Class imbalance is a common problem that has been comprehensively studied in classical machine learning, yet very limited systematic research is available in the context of deep learning. In our study, we use three benchmark datasets of increasing complexity, MNIST, CIFAR-10 and ImageNet, to investigate the effects of imbalance on classification and perform an extensive comparison of several methods to address the issue: oversampling, undersampling, two-phase training, and thresholding that compensates for prior class probabilities. Our main evaluation metric is area under the receiver operating characteristic curve (ROC AUC) adjusted to multi-class tasks since overall accuracy metric is associated with notable difficulties in the context of imbalanced data. Based on results from our experiments we conclude that (i) the effect of class imbalance on classification performance is detrimental; (ii) the method of addressing class imbalance that emerged as dominant in almost all analyzed scenarios was oversampling; (iii) oversampling should be applied to the level that completely eliminates the imbalance, whereas the optimal undersampling ratio depends on the extent of imbalance; (iv) as opposed to some classical machine learning models, oversampling does not cause overfitting of CNNs; (v) thresholding should be applied to compensate for prior class probabilities when overall number of properly classified cases is of interest. 

  • 9.
    Butepage, Judith
    et al.
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL. KTH Royal Inst Technol, CSC, Robot Percept & Learning Lab RPL, Stockholm, Sweden..
    Kjellström, Hedvig
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL. KTH Royal Inst Technol, CSC, Robot Percept & Learning Lab RPL, Stockholm, Sweden..
    Kragic, Danica
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL. KTH Royal Inst Technol, CSC, Robot Percept & Learning Lab RPL, Stockholm, Sweden..
    Anticipating many futures: Online human motion prediction and generation for human-robot interaction2018In: 2018 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA), IEEE COMPUTER SOC , 2018, p. 4563-4570Conference paper (Refereed)
    Abstract [en]

    Fluent and safe interactions of humans and robots require both partners to anticipate the others' actions. The bottleneck of most methods is the lack of an accurate model of natural human motion. In this work, we present a conditional variational autoencoder that is trained to predict a window of future human motion given a window of past frames. Using skeletal data obtained from RGB depth images, we show how this unsupervised approach can be used for online motion prediction for up to 1660 ms. Additionally, we demonstrate online target prediction within the first 300-500 ms after motion onset without the use of target specific training data. The advantage of our probabilistic approach is the possibility to draw samples of possible future motion patterns. Finally, we investigate how movements and kinematic cues are represented on the learned low dimensional manifold.

  • 10.
    Båberg, Fredrik
    et al.
    KTH, School of Computer Science and Communication (CSC), Robotics, perception and learning, RPL. KTH, School of Computer Science and Communication (CSC), Centres, Centre for Autonomous Systems, CAS.
    Petter, Ögren
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL. KTH, School of Electrical Engineering and Computer Science (EECS), Centres, Centre for Autonomous Systems, CAS.
    Formation Obstacle Avoidance using RRT and Constraint Based Programming2017In: 2017 IEEE International Symposium on Safety, Security and Rescue Robotics (SSRR), IEEE conference proceedings, 2017, article id 8088131Conference paper (Refereed)
    Abstract [en]

    In this paper, we propose a new way of doing formation obstacle avoidance using a combination of Constraint Based Programming (CBP) and Rapidly Exploring Random Trees (RRTs). RRT is used to select waypoint nodes, and CBP is used to move the formation between those nodes, reactively rotating and translating the formation to pass the obstacles on the way. Thus, the CBP includes constraints for both formation keeping and obstacle avoidance, while striving to move the formation towards the next waypoint. The proposed approach is compared to a pure RRT approach where the motion between the RRT waypoints is done following linear interpolation trajectories, which are less computationally expensive than the CBP ones. The results of a number of challenging simulations show that the proposed approach is more efficient for scenarios with high obstacle densities.

  • 11.
    Caccamo, Sergio
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Enhancing geometric maps through environmental interactions2018Doctoral thesis, comprehensive summary (Other academic)
    Abstract [en]

    The deployment of rescue robots in real operations is becoming increasingly commonthanks to recent advances in AI technologies and high performance hardware. Rescue robots can now operate for extended period of time, cover wider areas andprocess larger amounts of sensory information making them considerably more usefulduring real life threatening situations, including both natural or man-made disasters.

    In this thesis we present results of our research which focuses on investigating ways of enhancing visual perception for Unmanned Ground Vehicles (UGVs) through environmental interactions using different sensory systems, such as tactile sensors and wireless receivers.

    We argue that a geometric representation of the robot surroundings built upon vision data only, may not suffice in overcoming challenging scenarios, and show that robot interactions with the environment can provide a rich layer of new information that needs to be suitably represented and merged into the cognitive world model. Visual perception for mobile ground vehicles is one of the fundamental problems in rescue robotics. Phenomena such as rain, fog, darkness, dust, smoke and fire heavily influence the performance of visual sensors, and often result in highly noisy data, leading to unreliable or incomplete maps.

    We address this problem through a collection of studies and structure the thesis as follow:Firstly, we give an overview of the Search & Rescue (SAR) robotics field, and discuss scenarios, hardware and related scientific questions.Secondly, we focus on the problems of control and communication. Mobile robotsrequire stable communication with the base station to exchange valuable information. Communication loss often presents a significant mission risk and disconnected robotsare either abandoned, or autonomously try to back-trace their way to the base station. We show how non-visual environmental properties (e.g. the WiFi signal distribution) can be efficiently modeled using probabilistic active perception frameworks based on Gaussian Processes, and merged into geometric maps so to facilitate the SAR mission. We then show how to use tactile perception to enhance mapping. Implicit environmental properties such as the terrain deformability, are analyzed through strategic glancesand touches and then mapped into probabilistic models.Lastly, we address the problem of reconstructing objects in the environment. Wepresent a technique for simultaneous 3D reconstruction of static regions and rigidly moving objects in a scene that enables on-the-fly model generation. Although this thesis focuses mostly on rescue UGVs, the concepts presented canbe applied to other mobile platforms that operates under similar circumstances. To make sure that the suggested methods work, we have put efforts into design of user interfaces and the evaluation of those in user studies.

  • 12.
    Caccamo, Sergio Salvatore
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Joint 3D Reconstruction of a Static Scene and Moving Objects2017In: Proceedings of the 2017International Conference on 3D Vision (3DV’17), 2017Conference paper (Other academic)
  • 13.
    Carvalho, J. Frederico
    et al.
    KTH. KTH, CAS, RPL, Royal Inst Technol, Stocholm, Sweden..
    Vejdemo-Johansson, Mikael
    CUNY Coll Staten Isl, Math Dept, Staten Isl, NY 10314 USA.;CUNY, Grad Ctr, Comp Sci, New York, NY USA..
    Kragic, Danica
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL. KTH, CAS, RPL, Royal Inst Technol, Stocholm, Sweden..
    Pokorny, Florian T.
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL. KTH, CAS, RPL, Royal Inst Technol, Stocholm, Sweden..
    Path Clustering with Homology Area2018In: 2018 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA), IEEE Computer Society, 2018, p. 7346-7353Conference paper (Refereed)
    Abstract [en]

    Path clustering has found many applications in recent years. Common approaches to this problem use aggregates of the distances between points to provide a measure of dissimilarity between paths which do not satisfy the triangle inequality. Furthermore, they do not take into account the topology of the space where the paths are embedded. To tackle this, we extend previous work in path clustering with relative homology, by employing minimum homology area as a measure of distance between homologous paths in a triangulated mesh. Further, we show that the resulting distance satisfies the triangle inequality, and how we can exploit the properties of homology to reduce the amount of pairwise distance calculations necessary to cluster a set of paths. We further compare the output of our algorithm with that of DTW on a toy dataset of paths, as well as on a dataset of real-world paths.

  • 14. Colledancise, Michele
    et al.
    Parasuraman, Ramviyas Nattanmai
    Petter, Ögren
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Learning of Behavior Trees for Autonomous Agents2018In: IEEE Transactions on Games, ISSN 2475-1502Article in journal (Refereed)
    Abstract [en]

    In this paper, we study the problem of automatically synthesizing a successful Behavior Tree (BT) in an a-priori unknown dynamic environment. Starting with a given set of behaviors, a reward function, and sensing in terms of a set of binary conditions, the proposed algorithm incrementally learns a switching structure in terms of a BT, that is able to handle the situations encountered. Exploiting the fact that BTs generalize And-Or-Trees and also provide very natural chromosome mappings for genetic pro- gramming, we combine the long term performance of Genetic Programming with a greedy element and use the And-Or analogy to limit the size of the resulting structure. Finally, earlier results on BTs enable us to provide certain safety guarantees for the resulting system. Using the testing environment Mario AI we compare our approach to alternative methods for learning BTs and Finite State Machines. The evaluation shows that the proposed approach generated solutions with better performance, and often fewer nodes than the other two methods.

  • 15.
    Cruciani, Silvia
    et al.
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Smith, Christian
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Integrating Path Planning and Pivoting2018Conference paper (Refereed)
  • 16.
    Cruciani, Silvia
    et al.
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Smith, Christian
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Kragic, Danica
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Hang, Kaiyu
    Dexterous Manipulation Graphs2018Conference paper (Refereed)
  • 17.
    Djikic, Addi
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Segmentation and Depth Estimation of Urban Road Using Monocular Camera and Convolutional Neural Networks2018Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    Deep learning for safe autonomous transport is rapidly emerging. Fast and robust perception for autonomous vehicles will be crucial for future navigation in urban areas with high traffic and human interplay.

    Previous work focuses on extracting full image depth maps, or finding specific road features such as lanes. However, in urban environments lanes are not always present, and sensors such as LiDAR with 3D point clouds provide a quite sparse depth perception of road with demanding algorithmic approaches.

    In this thesis we derive a novel convolutional neural network that we call AutoNet. It is designed as an encoder-decoder network for pixel-wise depth estimation of an urban drivable free-space road, using only a monocular camera, and handled as a supervised regression problem. AutoNet is also constructed as a classification network to solely classify and segment the drivable free-space in real- time with monocular vision, handled as a supervised classification problem, which shows to be a simpler and more robust solution than the regression approach.

    We also implement the state of the art neural network ENet for comparison, which is designed for fast real-time semantic segmentation and fast inference speed. The evaluation shows that AutoNet outperforms ENet for every performance metrics, but shows to be slower in terms of frame rate. However, optimization techniques are proposed for future work, on how to advance the frame rate of the network while still maintaining the robustness and performance.

    All the training and evaluation is done on the Cityscapes dataset. New ground truth labels for road depth perception are created for training with a novel approach of fusing pre-computed depth maps with semantic labels. Data collection with a Scania vehicle is conducted, mounted with a monocular camera to test the final derived models.

    The proposed AutoNet shows promising state of the art performance in regards to road depth estimation as well as road classification.

  • 18.
    Ericson, Ludvig
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Flying High: Deep Imitation Learning of Optimal Control for Unmanned Aerial Vehicles2018Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    Optimal control for multicopters is difficult in part due to the low processing power available, and the instability inherent to multicopters. Deep imitation learning is a method for approximating an expert control policy with a neural network, and has the potential of improving control for multicopters. We investigate the performance and reliability of deep imitation learning with trajectory optimization as the expert policy by first defining a dynamics model for multicopters and applying a trajectory optimization algorithm to it. Our investigation shows that network architecture plays an important role in the characteristics of both the learning process and the resulting control policy, and that in particular trajectory optimization can be leveraged to improve convergence times for imitation learning. Finally, we identify some limitations and future areas of study and development for the technology.

  • 19.
    Ghadirzadeh, Ali
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Sensorimotor Robot Policy Training using Reinforcement Learning2018Doctoral thesis, comprehensive summary (Other academic)
    Abstract [en]

    Robots are becoming more ubiquitous in our society and taking over many tasks that were previously considered as human hallmarks. Many of these tasks, e.g., autonomously driving a car, collaborating with humans in dynamic and changing working conditions and performing household chores, require human-level intelligence to perceive the world and to act appropriately. In this thesis, we pursue a different approach compared to classical methods that often construct a robot controller based on the perception-then-action paradigm. We devise robotic action-selection policies by considering action-selection and perception processes as being intertwined, emphasizing that perception comes prior to action and action is key to perception. The main hypothesis is that complex robotic behaviors come as the result of mastering sensorimotor contingencies (SMCs), i.e., regularities between motor actions and associated changes in sensory observations, where SMCs can be seen as building blocks to skillful behaviors. We elaborate and investigate this hypothesis by deliberate design of frameworks which enable policy training merely based on data experienced by a robot,without intervention of human experts for analytical modelings or calibrations. In such circumstances, action policies can be obtained by reinforcement learning (RL) paradigm by making exploratory action decisions and reinforcing patterns of SMCs that lead to reward events for a given task. However, the dimensionality of sensorimotor spaces, complex dynamics of physical tasks, sparseness of reward events, limited amount of data from real-robot experiments, ambiguities of crediting past decisions and safety issues, which arise from exploratory actions of a physical robot, pose challenges to obtain a policy based on data-driven methods alone. In this thesis, we introduce our contributions to deal with the aforementioned issues by devising learning frameworks which endow a robot with the ability to integrate sensorimotor data to obtain action-selection policies. The effectiveness of the proposed frameworks is demonstrated by evaluating the methods on a number of real robotic tasks and illustrating the suitability of the methods to acquire different skills, to make sequential action-decisions in high-dimensional sensorimotor spaces, with limited data and sparse rewards.

  • 20.
    Guin, Agneev
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Terrain Classification to find Drivable Surfaces using Deep Neural Networks: Semantic segmentation for unstructured roads combined with the use of Gabor filters to determine drivable regions trained on a small dataset2018Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    Autonomous vehicles face various challenges under difficult terrain conditions such as marginally rural or back-country roads, due to the lack of lane information, road signs or traffic signals. In this thesis, we investigate a novel approach of using Deep Neural Networks (DNNs) to classify off-road surfaces into the types of terrains with the aim of supporting autonomous navigation in unstructured environments. For example, off-road surfaces can be classified as asphalt, gravel, grass, mud, snow, etc.

    Images from the camera mounted on a mining truck were used to perform semantic segmentation and to classify road surface types. Camera images were segmented manually for training into sets of 16 and 9 classes, for all relevant classes and the drivable classes respectively. A small but diverse dataset of 100 images was augmented and compiled along with nearby frames from the video clips to expand this dataset. Neural networks were used to test the performance for the classification under these off-road conditions. Pre-trained AlexNet was compared to the networks without pre-training. Gabor filters, known to distinguish textured surfaces, was further used to improve the results of the neural network.

    The experiments show that pre-trained networks perform well with small datasets and many classes. A combination of Gabor filters with pre-trained networks can establish a dependable navigation path under difficult terrain conditions. While the results seem positive for images similar to the training image scenes, the networks fail to perform well in other situations. Though the tests imply that larger datasets are required for dependable results, this is a step closer to making the autonomous vehicles drivable under off-road conditions.

  • 21.
    Guo, Meng
    et al.
    KTH, School of Electrical Engineering and Computer Science (EECS), Centres, ACCESS Linnaeus Centre. KTH, School of Electrical Engineering and Computer Science (EECS), Centres, Centre for Autonomous Systems, CAS.
    Boskos, Dimitris
    KTH, School of Electrical Engineering and Computer Science (EECS), Centres, ACCESS Linnaeus Centre. KTH, School of Electrical Engineering and Computer Science (EECS), Centres, Centre for Autonomous Systems, CAS.
    Tumova, Jana
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Dimarogonas, Dimos V.
    KTH, School of Electrical Engineering and Computer Science (EECS), Centres, ACCESS Linnaeus Centre. KTH, School of Electrical Engineering and Computer Science (EECS), Centres, Centre for Autonomous Systems, CAS.
    Distributed hybrid control synthesis for multi-agent systems from high-level specifications2018In: Control Subject to Computational and Communication Constraints, Springer Verlag , 2018, 475, p. 241-260Chapter in book (Refereed)
    Abstract [en]

    Current control applications necessitate in many cases the consideration of systems with multiple interconnected components. These components/agents may need to fulfill high-level tasks at a discrete planning layer and also coupled constraints at the continuous control layer. Toward this end, the need for combined decentralized control at the continuous layer and planning at the discrete layer becomes apparent. While there are approaches that handle the problem in a top-down centralized manner, decentralized bottom-up approaches have not been pursued to the same extent. We present here some of our results for the problem of combined, hybrid control and task planning from high-level specifications for multi-agent systems in a bottom-up manner. In the first part, we present some initial results on extending the necessary notion of abstractions to multi-agent systems in a distributed fashion. We then consider a setup where agents are assigned individual tasks in the form of linear temporal logic (LTL) formulas and derive local task planning strategies for each agent. In the last part, the problem of combined distributed task planning and control under coupled continuous constraints is further considered.

  • 22.
    Hamesse, Charles
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Simultaneous Measurement Imputation and Rehabilitation Outcome Prediction for Achilles Tendon Rupture2018Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    Achilles Tendon Rupture (ATR) is one of the typical soft tissue injuries. Rehabilitation after such musculoskeletal injuries remains a prolonged process with a very variable outcome. Being able to predict the rehabilitation outcome accurately is crucial for treatment decision support. In this work, we design a probabilistic model to predict the rehabilitation outcome for ATR using a clinical cohort with numerous missing entries. Our model is trained end-to-end in order to simultaneously predict the missing entries and the rehabilitation outcome. We evaluate our model and compare with multiple baselines, including multi-stage methods. Experimental results demonstrate the superiority of our model over these baseline multi-stage approaches with various data imputation methods for ATR rehabilitation outcome prediction.

  • 23.
    Hamesse, Charles
    et al.
    KTH.
    Ackermann, P.
    Kjellström, Hedvig
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Zhang, C.
    Simultaneous measurement imputation and outcome prediction for achilles tendon rupture rehabilitation2018In: CEUR Workshop Proceedings, CEUR-WS , 2018, Vol. 2142, p. 82-86Conference paper (Refereed)
    Abstract [en]

    Achilles Tendon Rupture (ATR) is one of the typical soft tissue injuries. Accurately predicting the rehabilitation outcome of ATR using noisy measurements with missing entries is crucial for treatment decision support. In this work, we design a probabilistic model that simultaneously predicts the missing measurements and the rehabilitation outcome in an end-to-end manner. We evaluate our model and compare it with multiple baselines including multi-stage methods using an ATR clinical cohort. Experimental results demonstrate the superiority of our model for ATR rehabilitation outcome prediction.

  • 24.
    Karlsson, Jesper
    et al.
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL. KTH Royal Inst Technol, Stockholm, Sweden..
    Vasile, Cristian-Ioan
    MIT, Cambridge, MA 02139 USA..
    Tumova, Jana
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL. KTH Royal Inst Technol, Stockholm, Sweden..
    Karaman, Sertac
    MIT, Cambridge, MA 02139 USA..
    Rus, Daniela
    MIT, Cambridge, MA 02139 USA..
    Multi-vehicle motion planning for social optimal mobility-on-demand2018In: 2018 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA), IEEE COMPUTER SOC , 2018, p. 7298-7305Conference paper (Refereed)
    Abstract [en]

    In this paper we consider a fleet of self-driving cars operating in a road network governed by rules of the road, such as the Vienna Convention on Road Traffic, providing rides to customers to serve their demands with desired deadlines. We focus on the associated motion planning problem that tradesoff the demands' delays and level of violation of the rules of the road to achieve social optimum among the vehicles. Due to operating in the same environment, the interaction between the cars must be taken into account, and can induce further delays. We propose an integrated route and motion planning approach that achieves scalability with respect to the number of cars by resolving potential collision situations locally within so-called bubble spaces enclosing the conflict. The algorithms leverage the road geometries, and perform joint planning only for lead vehicles in the conflict and use queue scheduling for the remaining cars. Furthermore, a framework for storing previously resolved conflict situations is proposed, which can be use for quick querying of joint motion plans. We show the mobility-on-demand setup and effectiveness of the proposed approach in simulated case studies involving up to 10 selfdriving vehicles.

  • 25.
    Kolibacz, Eric
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Classification of incorrectly picked components using Convolutional Neural Networks2018Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    Printed circuit boards used in most ordinary electrical devices are usually equipped through an assembly line. Pick and place machines as part of those lines require accurate detection of incorrectly picked components, and this is commonly performed via image analysis. The goal of this project is to investigate if we can achieve state-of-the-art performance in an industrial quality assurance task through the application of artificial neural networks. Experiments regarding different network architectures and data modifications are conducted to achieve precise image classification. Although the classification rates do not surpass or equal the rates of the existing vision-based detection system, there remains great potential in the deployment of a machine-learning-based algorithm into pick and place machines.

  • 26.
    Kontogiorgos, Dimosthenis
    et al.
    KTH, School of Electrical Engineering and Computer Science (EECS), Speech, Music and Hearing, TMH.
    Sibirtseva, Elena
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Pereira, André
    KTH, School of Electrical Engineering and Computer Science (EECS), Speech, Music and Hearing, TMH.
    Skantze, Gabriel
    KTH, School of Electrical Engineering and Computer Science (EECS), Speech, Music and Hearing, TMH.
    Gustafson, Joakim
    KTH, School of Electrical Engineering and Computer Science (EECS), Speech, Music and Hearing, TMH.
    Multimodal reference resolution in collaborative assembly tasks2018Conference paper (Refereed)
  • 27.
    Kragic, Danica
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL. KTH Royal Inst Technol, Ctr Autonomous Syst, Stockholm, Sweden.;KTH Royal Inst Technol, Comp Sci, Stockholm, Sweden..
    From active perception to deep learning2018In: SCIENCE ROBOTICS, ISSN 2470-9476, Vol. 3, no 23, article id eaav1778Article in journal (Other academic)
  • 28.
    Krug, Robert
    et al.
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL. KTH Royal Inst Technol, Robot Learning & Percept lab, S-10044 Stockholm, Sweden..
    Bekiroglu, Yasemin
    Vicarious AI, San Francisco, CA USA..
    Kragic, Danica
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL. KTH Royal Inst Technol, Robot Learning & Percept lab, S-10044 Stockholm, Sweden..
    Roa, Maximo A.
    German Aerosp Ctr DLR, Inst Robot & Mechatron, D-82234 Wessling, Germany..
    Evaluating the Quality of Non-Prehensile Balancing Grasps2018In: 2018 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA), IEEE Computer Society, 2018, p. 4215-4220Conference paper (Refereed)
    Abstract [en]

    Assessing grasp quality and, subsequently, predicting grasp success is useful for avoiding failures in many autonomous robotic applications. In addition, interest in non-prehensile grasping and manipulation has been growing as it offers the potential for a large increase in dexterity. However, while force-closure grasping has been the subject of intense study for many years, few existing works have considered quality metrics for non-prehensile grasps. Furthermore, no studies exist to validate them in practice. In this work we use a real-world data set of non-prehensile balancing grasps and use it to experimentally validate a wrench-based quality metric by means of its grasp success prediction capability. The overall accuracy of up to 84% is encouraging and in line with existing results for force-closure grasps.

  • 29.
    Kucherenko, Taras
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Data Driven Non-Verbal Behavior Generation for Humanoid Robots2018Conference paper (Refereed)
    Abstract [en]

    Social robots need non-verbal behavior to make an interaction pleasant and efficient. Most of the models for generating non-verbal behavior are rule-based and hence can produce a limited set of motions and are tuned to a particular scenario. In contrast, datadriven systems are flexible and easily adjustable. Hence we aim to learn a data-driven model for generating non-verbal behavior (in a form of a 3D motion sequence) for humanoid robots. Our approach is based on a popular and powerful deep generative model: Variation Autoencoder (VAE). Input for our model will be multi-modal and we will iteratively increase its complexity: first, it will only use the speech signal, then also the text transcription and finally - the non-verbal behavior of the conversation partner. We will evaluate our system on the virtual avatars as well as on two humanoid robots with different embodiments: NAO and Furhat. Our model will be easily adapted to a novel domain: this can be done by providing application specific training data.

  • 30.
    Lotz, Max
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Depth Inclusion for Classification and Semantic Segmentation2018Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    The  majority  of  computer  vision  algorithms  only  use  RGB  images  to  make  inferencesabout  the  state  of  the  world.  With  the  increasing  availability  of  RGB-D  cameras  it  is  im-portant  to  examine  ways  to  effectively  fuse  this  extra  modality  for  increased  effective-ness.  This  paper  examines  how  depth  can  be  fused  into  CNNs  to  increase  accuracy  in  thetasks  of  classification  and  semantic  segmentation,  as  well  as  examining  how  this  depthshould  best  be  effectively  encoded  prior  to  inclusion  in  the  network.  Concatenating  depthas  a  fourth  image  channel  and  modifying  the  dimension  of  the  initial  layer  of  a  pretrainedCNN  is  initially  examined.  Creating  a  separate  duplicate  network  to  train  depth  on,  andfusing  both  networks  in  later  stages  is  shown  to  be  an  effective  technique  for  both  tasks.The  results  show  that  depth  concatenation  is  an  ineffective  strategy  as  it  clamps  the  ac-curacy  to  the  lower  accuracy  of  the  two  modalities,  whilst  late  fusion  can  improve  thetask  accuracy  beyond  that  of  just  the  RGB  trained  network  for  both  tasks.  It  is  also  foundthat  methods  such  as  HHA  encoding  which  revolve  around  calculating  geometric  prop-erties  of  the  depth,  such  as  surface  normals,  are  a  superior  encoding  method  than  sim-pler  colour  space  transformations  such  as  HSV.  This  only  holds  true  when  these  depthimages  are  normalised  over  the  maximum  depth  of  the  dataset  as  opposed  to  the  maxi-mum  depth  of  each  individual  image,  thus  retaining  geometric  consistency  between  im-ages.  The  reverse  holds  true  for  simpler  colour  space  transformations.

  • 31.
    Mahler, Jeffrey
    et al.
    Univ Calif Berkeley, Dept Elect Engn & Comp Sci, Berkeley, CA 94705 USA..
    Pokorny, Florian T.
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Niyaz, Sherdil
    Univ Calif Berkeley, Dept Elect Engn & Comp Sci, Berkeley, CA 94705 USA..
    Goldberg, Ken
    Univ Calif Berkeley, Dept Elect Engn & Comp Sci, Berkeley, CA 94705 USA.;Univ Calif Berkeley, Dept Ind Engn & Operat Res, Berkeley, CA 94705 USA..
    Synthesis of Energy-Bounded Planar Caging Grasps Using Persistent Homology2018In: IEEE Transactions on Automation Science and Engineering, ISSN 1545-5955, E-ISSN 1558-3783, Vol. 15, no 3, p. 908-918Article in journal (Refereed)
    Abstract [en]

    For applications such as manufacturing, caging grasps restrict object motion without requiring complete immobilization, providing a robust alternative to force-and form-closure grasps. Energy-bounded cages are a new class of caging grasps that relax the requirement of complete caging in the presence of external forces such as gravity or constant velocity pushing in the horizontal plane with Coulomb friction. We address the problem of synthesizing planar energy-bounded cages by identifying gripper and force-direction configurations that maximize the energy required for the object to escape. We present Energy-Bounded-Cage-Synthesis-2-D (EBCS-2-D), a sampling-based algorithm that uses persistent homology, a recently-developed multiscale approach for topological analysis, to efficiently compute candidate rigid configurations of obstacles that form energy-bounded cages of an object from an alpha-shape approximation to the configuration space. If a synthesized configuration has infinite escape energy then the object is completely caged. EBCS-2-D runs in O(s(3) + sn(2)) time, where s is the number of samples and n is the number of object and obstacle vertices, where typically n << s. We observe runtimes closer to O(s) for fixed n. We implement EBCS-2-D using the persistent homology algorithms toolbox and study performance on a set of seven planar objects and four gripper types. Experiments suggest that EBCS-2-D takes 2-3 min on a 6 core processor with 200 000 pose samples. We also confirm that an rapidly-exploring random tree* motion planner is unable to find escape paths with lower energy. Physical experiments on a five degree of freedom Zymark Zymate and ABB YuMi suggest that push grasps synthesized by EBCS-2-D are robust to perturbations. Data and code are available at http://berkeleyautomation.github.io/caging/.

  • 32.
    Masud, Nauman
    et al.
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL. Univ Gavle, Dept Elect Math & Sci, S-80176 Gavle, Sweden.
    Smith, Christian
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Isaksson, Magnus
    Univ Gavle, Dept Elect Math & Sci, S-80176 Gavle, Sweden..
    Disturbance observer based dynamic load torque compensator for assistive exoskeletons2018In: Mechatronics (Oxford), ISSN 0957-4158, E-ISSN 1873-4006, Vol. 54, p. 78-93Article in journal (Refereed)
    Abstract [en]

    In assistive robotics applications, the human limb is attached intimately to the robotic exoskeleton. The coupled dynamics of the human-exoskeleton system are highly nonlinear and uncertain, and effectively appear as uncertain load-torques at the joint actuators of the exoskeleton. This uncertainty makes the application of standard computed torque techniques quite challenging. Furthermore, the need for safe human interaction severely limits the gear ratio of the actuators. With small gear ratios, the uncertain joint load-torques cannot be ignored and need to be effectively compensated. A novel disturbance observer based dynamic load-torque compensator is hereby proposed and analysed for the current controlled DC-drive actuators of the exoskeleton, to effectively compensate the said uncertain load-torques at the joint level. The feedforward dynamic load-torque compensator is proposed based on the higher order dynamic model of the current controlled DC-drive. The dynamic load-torque compensator based current controlled DC-drive is then combined with a tailored feedback disturbance observer to further improve the compensation performance in the presence of drive parametric uncertainty. The proposed compensator structure is shown both theoretically and practically to give significantly improved performance w.r.t disturbance observer compensator alone and classical static load-torque compensator, for rated load-torque frequencies up to 1.6 Hz, which is a typical joint frequency bound for normal daily activities for elderly. It is also shown theoretically that the proposed compensator achieves the improved performance with comparable reference current requirement for the current controlled DC-drive.

  • 33. Menghi, C.
    et al.
    García, S.
    Pelliccione, P.
    Tumova, Jana
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Poster: Towards multi-robot applications planning under uncertainty2018In: Proceedings of the 40th International Conference on Software Engineering: Companion Proceeedings, IEEE Computer Society, 2018, p. 438-439Conference paper (Refereed)
    Abstract [en]

    Novel robotic applications are no longer based on single robots. They rather require teams of robots that collaborate and interact to perform a desired mission. They must also be used in contexts in which only partial knowledge about the robots and their environment is present. To ensure mission achievement, robotic applications require the usage of planners that compute the set of actions the robots must perform. Current planning techniques are often based on centralized solutions and hence they do not scale when teams of robots are considered, they consider rather simple missions, and they do not work in partially known environments. To address these challenges, we present a planning solution that decomposes the team of robots into subclasses, considers complex high-level missions given in temporal logic, and at the same time works when only partial knowledge of the environment is available.

  • 34.
    Mikheeva, Olga
    et al.
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Ek, C. H.
    Kjellström, Hedvig
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Perceptual facial expression representation2018In: Proceedings - 13th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2018, Institute of Electrical and Electronics Engineers (IEEE), 2018, p. 179-186Conference paper (Refereed)
    Abstract [en]

    Dissimilarity measures are often used as a proxy or a handle to reason about data. This can be problematic, as the data representation is often a consequence of the capturing process or how the data is visualized, rather than a reflection of the semantics that we want to extract. Facial expressions are a subtle and essential part of human communication but they are challenging to extract from current representations. In this paper we present a method that is capable of learning semantic representations of faces in a data driven manner. Our approach uses sparse human supervision which our method grounds in the data. We provide experimental justification of our approach showing that our representation improves the performance for emotion classification.

  • 35.
    Möckelind, Christoffer
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Improving deep monocular depth predictions using dense narrow field of view depth images2018Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    In this work we study a depth prediction problem where we provide a narrow field of view depth image and a wide field of view RGB image to a deep network tasked with predicting the depth for the entire RGB image. We show that by providing a narrow field of view depth image, we improve results for the area outside the provided depth compared to an earlier approach only utilizing a single RGB image for depth prediction. We also show that larger depth maps provide a greater advantage than smaller ones and that the accuracy of the model decreases with the distance from the provided depth. Further, we investigate several architectures as well as study the effect of adding noise and lowering the resolution of the provided depth image. Our results show that models provided low resolution noisy data performs on par with the models provided unaltered depth.

  • 36.
    Nassir, Cesar
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Domain-Independent Moving Object Depth Estimation using Monocular Camera2018Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    Today automotive companies across the world strive to create vehicles with fully autonomous capabilities. There are many benefits of developing autonomous vehicles, such as reduced traffic congestion, increased safety and reduced pollution, etc. To be able to achieve that goal there are many challenges ahead, one of them is visual perception.

    Being able to estimate depth from a 2D image has been shown to be a key component for 3D recognition, reconstruction and segmentation. Being able to estimate depth in an image from a monocular camera is an ill-posed problem since there is ambiguity between the mapping from colour intensity and depth value. Depth estimation from stereo images has come far compared to monocular depth estimation and was initially what depth estimation relied on. However, being able to exploit monocular cues is necessary for scenarios when stereo depth estimation is not possible.

    We have presented a novel CNN network, BiNet which is inspired by ENet, to tackle depth estimation of moving objects using only a monocular camera in real-time. It performs better than ENet in the Cityscapes dataset while adding only a small overhead to the complexity.

  • 37.
    Nilsson, Mårten
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Augmenting High-Dimensional Data with Deep Generative Models2018Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    Data augmentation is a technique that can be performed in various ways to improve the training of discriminative models. The recent developments in deep generative models offer new ways of augmenting existing data sets. In this thesis, a framework for augmenting annotated data sets with deep generative models is proposed together with a method for quantitatively evaluating the quality of the generated data sets. Using this framework, two data sets for pupil localization was generated with different generative models, including both well-established models and a novel model proposed for this purpose. The unique model was shown both qualitatively and quantitatively to generate the best data sets. A set of smaller experiments on standard data sets also revealed cases where this generative model could improve the performance of an existing discriminative model. The results indicate that generative models can be used to augment or replace existing data sets when training discriminative models.

  • 38.
    Nordström, Marcus
    et al.
    KTH, School of Engineering Sciences (SCI), Mathematics (Dept.), Mathematical Statistics.
    Hult, Henrik
    KTH, School of Engineering Sciences (SCI), Mathematics (Dept.), Mathematical Statistics.
    Maki, Atsuto
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Löfman, Fredrik
    Raysearch Labs, Stockholm, Sweden..
    Pareto Dose Prediction Using Fully Convolutional Networks Operating in 3D2018In: Medical physics (Lancaster), ISSN 0094-2405, Vol. 45, no 6, p. E176-E176Article in journal (Other academic)
  • 39.
    Pinto Basto de Carvalho, Joao Frederico
    et al.
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Vejdemo-Johansson, Mikael
    CUNY, Math Dept, Coll Staten Isl, New York, NY 10021 USA..
    Kragic, Danica
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Pokorny, Florian T.
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    An algorithm for calculating top-dimensional bounding chains2018In: PEERJ COMPUTER SCIENCE, ISSN 2376-5992, article id e153Article in journal (Refereed)
    Abstract [en]

    We describe the Coefficient-Flow algorithm for calculating the bounding chain of an (n-1)-boundary on an n-manifold-like simplicial complex S. We prove its correctness and show that it has a computational time complexity of O(vertical bar S(n-1)vertical bar) (where S(n-1) is the set of (n-1)-faces of S). We estimate the big-O coefficient which depends on the dimension of S and the implementation. We present an implementation, experimentally evaluate the complexity of our algorithm, and compare its performance with that of solving the underlying linear system.

  • 40.
    Potuaud, Sylvain
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Human Grasp Synthesis with Deep Learning2018Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    The human hands are one of the most complex organs of the human body. As they enable us to grasp various objects in many different ways, they have played a crucial role in the rise of the human race. Being able to control hands as a human do is a key step towards friendly human-robots interaction and realistic virtual human simulations.

    Grasp generation has been mostly studied for the purpose of generating physically stable grasps. This paper addresses a different aspect: how to generate realistic, natural looking grasps that are similar to human grasps. To simplify the problem, the wrist position is assumed to be known and only the finger pose is generated.

    As the realism of a grasp is not easy to put into equations, data-driven machine learning techniques are used. This paper investigated the application of the deep neural networks to the grasp generation problems. Two different object shape representations (point cloud and multi-view depth images), and multiple network architectures are experimented, using a collected human grasping dataset in a virtual reality environment.

    The resulting generated grasps are highly realistic and human-like. Though there are sometimes some finger penetrations on the object surface, the general poses of the fingers around the grasped objects are similar to the collected human data. The good performance extends to the objects of categories previously unseen by the network.

    This work has validated the efficiency of a data-driven deep learning approach for human-like grasp synthesis. I believe the realistic-looking objective of the grasp synthesis investigated in this thesis can be combined with the existing mechanical, stable grasp criteria to achieve both natural-looking and reliable grasp generations.

  • 41.
    Rai, Akshara
    et al.
    Carnegie Mellon Univ, Sch Comp Sci, Robot Inst, Pittsburgh, PA 15213 USA..
    Antonova, Rika
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL. KTH Royal Inst Technol, CSC, Robot Percept & Learning, Stockholm, Sweden..
    Song, Seungmoon
    Carnegie Mellon Univ, Sch Comp Sci, Robot Inst, Pittsburgh, PA 15213 USA..
    Martin, William
    Carnegie Mellon Univ, Sch Comp Sci, Robot Inst, Pittsburgh, PA 15213 USA..
    Geyer, Hartmut
    Carnegie Mellon Univ, Sch Comp Sci, Robot Inst, Pittsburgh, PA 15213 USA..
    Atkeson, Christopher
    Carnegie Mellon Univ, Sch Comp Sci, Robot Inst, Pittsburgh, PA 15213 USA..
    Bayesian Optimization Using Domain Knowledge on the ATRIAS Biped2018In: 2018 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA), IEEE, 2018, p. 1771-1778Conference paper (Refereed)
    Abstract [en]

    Robotics controllers often consist of expert-designed heuristics, which can be hard to tune in higher dimensions. Simulation can aid in optimizing these controllers if parameters learned in simulation transfer to hardware. Unfortunately, this is often not the case in legged locomotion, necessitating learning directly on hardware. This motivates using data-efficient learning techniques like Bayesian Optimization (BO) to minimize collecting expensive data samples. BO is a black-box data-efficient optimization scheme, though its performance typically degrades in higher dimensions. We aim to overcome this problem by incorporating domain knowledge, with a focus on bipedal locomotion. In our previous work, we proposed a feature transformation that projected a 16-dimensional locomotion controller to a 1-dimensional space using knowledge of human walking. When optimizing a human-inspired neuromuscular controller in simulation, this feature transformation enhanced sample efficiency of BO over traditional BO with a Squared Exponential kernel. In this paper, we present a generalized feature transform applicable to non-humanoid robot morphologies and evaluate it on the ATRIAS bipedal robot, in both simulation and hardware. We present three different walking controllers and two are evaluated on the real robot. Our results show that this feature transform captures important aspects of walking and accelerates learning on hardware and simulation, as compared to traditional BO.

  • 42.
    Rasines Suárez, Javier
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Gaussian process-assisted frontier exploration and indoor radio source localization for mobile robots2018Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    Autonomous localization of a radio source is addressed, in the context of autonomous charging for drones in indoor environments. A radio beacon will be the only input used by the robot to navigate to an unknown charging station, at an unknown area. Previous proposed algorithms used frontier-based exploration and the measured RSS to compute the direction to the source. The use of Gaussian processes is studied to model the Radio Signal Strength (RSS) distribution and generate an estimation of the gradient. This gradient was also incorporated into a frontier exploration algorithm and was compared with the proposed algorithm. It was found that the usefulness of the Gaussian process model depended on the distribution of the RSS samples. If the robot had no prior samples of the RSS, then the gradient-assisted solution performed better. Instead, if the robot had some prior knowledge of the RSS distribution, then the Gaussian process model yields a better performance.

  • 43.
    Sibirtseva, Elena
    et al.
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Kontogiorgos, Dimosthenis
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Nykvist, Olov
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Karaoguz, Hakan
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Leite, Iolanda
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Gustafson, Joakim
    KTH, School of Electrical Engineering and Computer Science (EECS), Speech, Music and Hearing, TMH.
    Kragic, Danica
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    A Comparison of Visualisation Methods for Disambiguating Verbal Requests in Human-Robot Interaction2018In: 2018 27th IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN) 2018 27th IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN), 2018Conference paper (Refereed)
    Abstract [en]

    Picking up objects requested by a human user is a common task in human-robot interaction. When multiple objects match the user's verbal description, the robot needs to clarify which object the user is referring to before executing the action. Previous research has focused on perceiving user's multimodal behaviour to complement verbal commands or minimising the number of follow up questions to reduce task time. In this paper, we propose a system for reference disambiguation based on visualisation and compare three methods to disambiguate natural language instructions. In a controlled experiment with a YuMi robot, we investigated realtime augmentations of the workspace in three conditions - head-mounted display, projector, and a monitor as the baseline - using objective measures such as time and accuracy, and subjective measures like engagement, immersion, and display interference. Significant differences were found in accuracy and engagement between the conditions, but no differences were found in task time. Despite the higher error rates in the head-mounted display condition, participants found that modality more engaging than the other two, but overall showed preference for the projector condition over the monitor and head-mounted display conditions.

  • 44.
    Stefansson, Thor
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    3D obstacle avoidance for drones using a realistic sensor setup2018Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    Obstacle avoidance is a well researched area, however most of the works only consider a 2D environment. Drones can move in three dimensions. It is therefore of interest to develop a system that ensures safe flight in these three dimensions. Obstacle avoidance is of highest importance for drones if they are intended to work autonomously and around humans, since drones are often fragile and have fast moving propellers that can hurt humans. This project is based on the obstacle restriction algorithm in 3D, and uses OctoMap to conveniently use the sensor data from multiple sensors simultaneously and to deal with their limited field of view. The results show that the system is able to avoid obstacles in 3D.

  • 45.
    Tardioli, Danilo
    et al.
    University of Zaragoza.
    Parasuraman, Ramviyas
    University of Georgia.
    Ögren, Petter
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Pound: A multi-master ROS node for reducing delay and jitter in wireless multi-robot networks2019In: Robotics and Autonomous Systems, ISSN 0921-8890, E-ISSN 1872-793X, Vol. 111, p. 73-87Article in journal (Refereed)
    Abstract [en]

    The Robot Operating System (ROS) is a popular and widely used software framework for building robotics systems. With the growth of its popularity, it has started to be used in multi-robot systems as well. However, the TCP connections that the platform relies on for connecting the so-called ROS nodes presents several issues regarding limited-bandwidth, delays, and jitter, when used in wireless multi-hop networks. In this paper, we present a thorough analysis of the problem and propose a new ROS node called Pound to improve the wireless communication performance by reducing delay and jitter in data exchanges, especially in multi-hop networks. Pound allows the use of multiple ROS masters (roscores), features data compression, and importantly, introduces a priority scheme that allows favoring more important flows over less important ones. We compare Pound to the state-of-the-art solutions through extensive experiments and show that it performs equally well, or better in all the test cases, including a control-over-network example.

  • 46.
    Teye, Mattias
    et al.
    KTH, School of Electrical Engineering and Computer Science (EECS).
    Azizpour, Hossein
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Smith, Kevin
    KTH, School of Electrical Engineering and Computer Science (EECS), Computational Science and Technology (CST). Science for Life Laboratory.
    Bayesian Uncertainty Estimation for Batch Normalized Deep Networks2018Conference paper (Refereed)
    Abstract [en]

    We show that training a deep network using batch normalization is equivalent to approximate inference in Bayesian models. We further demonstrate that this finding allows us to make meaningful estimates of the model uncertainty using conventional architectures, without modifications to the network or the training procedure. Our approach is thoroughly validated by measuring the quality of uncertainty in a series of empirical experiments on different tasks. It outperforms baselines with strong statistical significance, and displays competitive performance with recent Bayesian approaches

  • 47.
    Trichon, Vincent
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    A Singing Drone Choir2017Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    Drones have a new emerging use case: performing in shows and live events. This master thesis has been driven by an artistic project invited to take part in a full-scale operatic performance in the Croatian National Theatre Ivan Zajc in Rijeka, Croatia, in 2019. This project merges technological research with ancient theatrical and operatic traditions by using drones as an opera choir. After describing the process of designing and building a fleet of quadrotors equipped with speakers, we present a reacting and interacting motion planning strategy based on potential fields. We analyse and evaluate our drone design with its control strategy on simulation and on a real drone.

  • 48.
    Vijayan, Aravind Elanjimattathil
    et al.
    KTH. KTH Royal Inst Technol, Sch Elect Engn & Comp Sci, Stockholm, Sweden..
    Alexanderson, Simon
    KTH. KTH Royal Inst Technol, Sch Elect Engn & Comp Sci, Stockholm, Sweden..
    Beskow, Jonas
    KTH, School of Electrical Engineering and Computer Science (EECS), Speech, Music and Hearing, TMH. KTH Royal Inst Technol, Sch Elect Engn & Comp Sci, Stockholm, Sweden..
    Leite, Iolanda
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL. KTH Royal Inst Technol, Sch Elect Engn & Comp Sci, Stockholm, Sweden..
    Using Constrained Optimization for Real-Time Synchronization of Verbal and Nonverbal Robot Behavior2018In: 2018 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA), IEEE Computer Society, 2018, p. 1955-1961Conference paper (Refereed)
    Abstract [en]

    Most of the motion re-targeting techniques are grounded on virtual character animation research, which means that they typically assume that the target embodiment has unconstrained joint angular velocities. However, because robots often do have such constraints, traditional re-targeting approaches can originate irregular delays in the robot motion. With the goal of ensuring synchronization between verbal and nonverbal behavior, this paper proposes an optimization framework for processing re-targeted motion sequences that addresses constraints such as joint angle and angular velocities. The proposed framework was evaluated on a humanoid robot using both objective and subjective metrics. While the analysis of the joint motion trajectories provides evidence that our framework successfully performs the desired modifications to ensure verbal and nonverbal behavior synchronization, results from a perceptual study showed that participants found the robot motion generated by our method more natural, elegant and lifelike than a control condition.

  • 49.
    Ward, Erik
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Models Supporting Trajectory Planning in Autonomous Vehicles2018Doctoral thesis, comprehensive summary (Other academic)
    Abstract [en]

    Autonomous vehicles have the potential to drastically improve the safety, efficiency and cost of transportation. Instead of a driver, an autonomous vehicle is controlled by an algorithm, offering improved consistency and the potential to eliminate human error from driving: by far the most common cause of accidents.

    Data collected from different types of sensors, along with prior information such as maps, are used to build models of the surrounding traffic scene, encoding relevant aspects of the driving problem.These models allow the autonomous vehicle to plan how it will drive, optimizing comfort, safety and progress towards its destination. To do so we must first encode the context of the current driving situation: the road geometry, where different traffic participants are, including the autonomous vehicle, and what routes are available to them. To plan the autonomous vehicle's trajectory, we also require models of how other traffic participants are likely to move in the near future, and what risks are incurred for different potential trajectories of the autonomous vehicle. In this thesis we present an overview of different trajectory planning approaches and the models enabling them along with our contributions towards localization, intention recognition, predictive behavior models and risk inference methods that support trajectory planning.

    Our first contribution is a method that allows localization of anautonomous vehicle using automotive short range radars. Furthermore, we investigate behavior recognition and prediction using models at two different levels of abstraction. We have also explored the integration of two different trajectory planning algorithms and probabilistic environment models which allow us to optimize the expected cost of chosen trajectories.

  • 50.
    Ward, Erik
    et al.
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL. KTH.
    Folkesson, John
    KTH, School of Electrical Engineering and Computer Science (EECS), Robotics, perception and learning, RPL.
    Towards Risk Minimizing Trajectory Planning in On-Road ScenariosManuscript (preprint) (Other academic)
    Abstract [en]

    Trajectory planning for autonomous vehicles should attempt to minimize expected risk given noisy sensor data and uncertain predictions of the near future. In this paper, we present a trajectory planning approach for on-road scenarios where we use a graph search approximation. Uncertain predictions of other vehicles are accounted for by anovel inference technique that allows efficient calculation of the probability of dangerous outcomes for set of modeled situation types.

12 1 - 50 of 52
CiteExportLink to result list
Permanent link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf