Reliable motion prediction of surrounding vehicles is the key to safe and efficient driving of autonomous vehicles, especially at urban intersections with complex traffic environments. This study models driving intentions and future driving speeds at urban intersections and improves model prediction performance by considering traffic environment characteristics. Key feature parameters including environmental characteristics are first extracted through driving behavior analysis and existing research experience. Then models with different input combinations are constructed to explore the effectiveness of different factors in predicting driving intention and future speed. In particular, in vehicle speed modeling, a target detection algorithm is used to identify traffic participants. Based on the identified traffic participant and vehicle position information, a new method for speed prediction that can reflect the dynamic interaction characteristics between the driver and the traffic environment is proposed. Models are trained and tested using natural driving data from China. Finally, the models with the simplest input and the best effect are determined. The driving intention recognition model can accurately predict the driving maneuvers of straight-Ahead, stopping, turning left and right 4 seconds before reaching the intersection. The speed prediction model can significantly improve the speed prediction accuracy, and shows stronger robustness and adaptability than existing models. This research provides important technical support for developing intelligent driving systems suitable for complex urban traffic environments.

Automated Vehicle (AV) validation based on simulated testing requires unbiased evaluation and high efficiency. One effective solution is to increase the exposure to risky rare events while reweighting the probability measure. However, characterizing the distribution of risky events is particularly challenging due to the paucity of samples and the temporality of continuous scenario variables. To solve it, we devise a method to represent, generate, and reweight the distribution of risky rare events. We decompose the temporal evolution of continuous variables into distribution components based on conditional probability. By introducing the Risk Indicator Function, the distribution of risky rare events is theoretically precipitated out of naturalistic driving distribution. This targeted distribution is practically generated via Normalizing Flow, which achieves exact and tractable probability evaluation of intricate distribution. The rare event distribution is then demonstrated as the advantageous Importance Sampling distribution. We also promote the technique of temporal Importance Sampling. The combined method, named as TrimFlow, is executed to estimate the collision rate of Car-following scenarios as a tentative practice. The results showed that sampling background vehicle maneuvers from rare event distribution could evolve testing scenarios to hazardous states. TrimFlow reduced 86.1% of tests compared to generating testing scenarios according to their exposure in the naturalistic driving environment. In addition, the TrimFlow method is not limited to one specific type of functional scenario.

Identification and classification of cell-graph features using graph-neural networks (GNNs) has been shown to be useful in digital pathology. In this work, we consider the role of edge labels in cell-graph modeling, including histological modeling techniques, edge aggregation in GNN architectures, and edge label prediction. We propose EAGNN (Edge Aggregated GNN), a new GNN model that aggregates both node and edge label information to take advantage of topological information about cellular data and facilitate edge label prediction. We introduce new edge label features that improve histological modeling and prediction. We evaluate our EAGNN model for the task of detecting the presence and location of the basement membrane in oral mucosal tissue, as a proof-of-concept application.

Background Histological feature representation is advantageous for computer aided diagnosis (CAD) and disease classification when using predictive techniques based on machine learning. Explicit feature representations in computer tissue models can assist explainability of machine learning predictions. Different approaches to feature representation within digital tissue images have been proposed. Cell-graphs have been demonstrated to provide precise and general constructs that can model both low- and high-level features. The basement membrane is high-level tissue architecture, and interactions across the basement membrane are involved in multiple disease processes. Thus, the basement membrane is an important histological feature to study from a cell-graph and machine learning perspective. Results We present a two stage machine learning pipeline for generating a cell-graph from a digital H &E stained tissue image. Using a combination of convolutional neural networks for visual analysis and graph neural networks exploiting node and edge labels for topological analysis, the pipeline is shown to predict both low- and high-level histological features in oral mucosal tissue with good accuracy. Conclusions Convolutional and graph neural networks are complementary technologies for learning, representing and predicting local and global histological features employing node and edge labels. Their combination is potentially widely applicable in histopathology image analysis and can enhance explainability in CAD tools for disease prediction.

Autonomous driving represents a significant challenge to all software quality assurance techniques, including testing. Generative machine learning (ML) techniques including active ML have considerable potential to generate high quality synthetic test data that can complement and improve on existing techniques such as hardware-in-the-loop and road testing.

As a methodology for system design and testing, use cases are well-known and widely used. While current active machine learning (ML) algorithms can effectively automate unit testing, they do not scale up to use case testing of complex systems in an efficient way. We present a new parallel distributed processing (PDP) architecture for a constrained active machine learning (CAML) approach to use case testing. To exploit CAML we introduce a use case modeling language with: (i) compile-time constraints on query generation, and (ii) run-time constraints using dynamic constraint checking. We evaluate this approach by applying a prototype implementation of CAML to use case testing of simulated multi-vehicle autonomous driving scenarios.

Background: Program similarity is a fundamental concept, central to the solution of software engineering tasks such as software plagiarism, clone identification, code refactoring and code search. Accurate similarity estimation between programs requires an in-depth understanding of their structure, semantics and flow. A control flow graph (CFG), is a graphical representation of a program which captures its logical control flow and hence its semantics. A common approach is to estimate program similarity by analysing CFGs using graph similarity measures, e.g. graph edit distance (GED). However, graph edit distance is an NP-hard problem and computationally expensive, making the application of graph similarity techniques to complex software programs impractical. Aim: This study intends to examine the effectiveness of graph neural networks to estimate program similarity, by analysing the associated control flow graphs. Method: We introduce funcGNN1, which is a graph neural network trained on labeled CFG pairs to predict the GED between unseen program pairs by utilizing an effective embedding vector. To our knowledge, this is the first time graph neural networks have been applied on labeled CFGs for estimating the similarity between highlevel language programs. Results: We demonstrate the effectiveness of funcGNN to estimate the GED between programs and our experimental analysis demonstrates how it achieves a lower error rate (1.94 ×10-3), with faster (23 times faster than the quickest traditional GED approximation method) and better scalability compared with state of the art methods. Conclusion: funcGNN posses the inductive learning ability to infer program structure and generalise to unseen programs. The graph embedding of a program proposed by our methodology could be applied to several related software engineering problems (such as code plagiarism and clone identification) thus opening multiple research directions.

We explore the application of graph database technology to spatio-temporal model checking of cooperating cyber-physical systems-of-systems such as vehicle platoons. We present a translation of spatio-temporal automata(STA) and the spatio-temporal logic STAL to se-mantically equivalent property graphs and graph queries respectively. We prove a sound reduction of the spatio-temporal verification problem tograph database query solving. The practicability and efficiency of thisapproach is evaluated by introducing NeoMC, a prototype implementation of our explicit model checking approach based on Neo4j. To evaluate NeoMC we consider case studies of verifying vehicle platooning models. Our evaluation demonstrates the effectiveness of our approach in terms of execution time and counterexample detection.

An oracle is used in software testing to derive the verdict (pass/fail) for a test case. Lack of precise test oracles is one of the major problems in software testing which can hinder judgements about quality. Metamorphic testing is an emerging technique which solves both the oracle problem and the test case generation problem by testing special forms of software requirements known as metamorphic requirements. However, manually deriving the metamorphic requirements for a given program requires a high level of domain expertise, is labor intensive and error prone. As an alternative, we consider the problem of automatic detection of metamorphic requirements using machine learning (ML). For this problem we can apply graph kernels and support vector machines (SVM). A significant problem for any ML approach is to obtain a large labeled training set of data (in this case programs) that generalises well. The main contribution of this paper is a general method to generate large volumes of synthetic training data which can improve ML assisted detection of metamorphic requirements. For training data synthesis we adopt mutation testing techniques. This research is the first to explore the area of data augmentation techniques for ML-based analysis of software code. We also have the goal to enhance black-box testing using white-box methodologies. Our results show that the mutants incorporated into the source code corpus not only efficiently scale the dataset size, but they can also improve the accuracy of classification models.

Cooperating cyber-physical systems-of-systems (CO-CPS) such as vehicle platoons, robot teams or drone swarms usually have strict safety requirements on both spatial and temporal behavior. Learning-based testing is a combination of machine learning and model checking that has been successfully used for black-box requirements testing of cyber-physical systems-of-systems. We present an overview of research in progress to apply learning-based testing to evaluate spatio-temporal requirements on autonomous systems-of-systems through modeling and simulation.