kth.sePublications KTH
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • 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
Transitional Grid Maps for Adverse Weather Conditions: Temporal Sparsity-based Filtering on Range Sensors
KTH, School of Industrial Engineering and Management (ITM), Engineering Design, Mechatronics and Embedded Control Systems.ORCID iD: 0000-0001-9982-578X
KTH, School of Electrical Engineering and Computer Science (EECS), Intelligent systems, Robotics, Perception and Learning, RPL.ORCID iD: 0000-0001-8747-6359
KTH, School of Electrical Engineering and Computer Science (EECS), Intelligent systems, Robotics, Perception and Learning, RPL.ORCID iD: 0000-0003-4173-2593
KTH, School of Electrical Engineering and Computer Science (EECS), Intelligent systems, Robotics, Perception and Learning, RPL.ORCID iD: 0000-0002-1170-7162
Show others and affiliations
(English)Manuscript (preprint) (Other academic)
Abstract [en]

Autonomous agents rely on accurate environment perception for safe and effective navigation. However, adverse weather conditions, such as snow or heavy rain, pose significant challenges to range sensors, leading to false positives and hindered obstacle detection. Previous work has focused on analyzing the sparsity of the measurements and learning from extensive datasets to filter out adverse weather detections. Despite promising results, the small residual unfiltered detections can still severely limit an agent's ability to generate feasible motion plans. In this paper, we introduce Transitional Grid Maps for Adverse Weather (TGMw). A temporal filter designed to distinguish weather-related noise from dynamic and static obstacles. It can either be used to directly filter the raw sensor measurements or to complement existing spatial filters. We demonstrate the benefits of our method in an extensive analysis in combination with multiple existing filters.

National Category
Robotics and automation
Identifiers
URN: urn:nbn:se:kth:diva-363918OAI: oai:DiVA.org:kth-363918DiVA, id: diva2:1961475
Funder
Vinnova, 876038
Note

QC 20250527

Available from: 2025-05-27 Created: 2025-05-27 Last updated: 2025-05-27Bibliographically approved
In thesis
1. Situation Awareness for Autonomous Agents under Limited Sensing
Open this publication in new window or tab >>Situation Awareness for Autonomous Agents under Limited Sensing
2025 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

Autonomous agents, such as robots and automated vehicles, rely on their ability to perceive and interpret their environment to make informed decisions and execute actions that align with their goals. A key aspect of this capability is situation awareness, which involves understanding the current state of the environment and predicting its future evolution. Traditional autonomous systems address perception and prediction as separate tasks within a sequential pipeline, where raw sensor data is processed into increasingly abstract representations. While this structured approach has driven significant advancements, it remains constrained by sensor limitations, including occlusions, measurement uncertainty, and adverse weather conditions.

This thesis investigates how predictions from past observations can enhance perception algorithms, enabling agents to infer missing information, reduce uncertainty, and better anticipate changes. To support this integration, alternative environment representations are explored that allow feedback between prediction and perception while capturing uncertainty. This tighter coupling improves decision-making, particularly in complex and partially observable environments.

The contributions include: (1) a reachability-based reasoning framework for tracking possible hidden obstacles; (2) its extension to handle delayed and partial external data; (3) a probabilistic mapping method, Transitional Grid Maps (TGM), that jointly models static and dynamic occupancy; and (4) an extension of TGM to mitigate weather-induced sensor noise.

The proposed methods are evaluated in simulated and real scenarios where traditional perception pipelines struggle, such as occluded, highly dynamic and noisy environments. By bridging the gap between perception and prediction, this work contributes to the development of more robust and intelligent autonomous systems.

Abstract [sv]

Autonoma agenter, såsom robotar och självkörande fordon, är beroende av sin förmåga att uppfatta och tolka omgivningen för att fatta välgrundade beslut och utföra handlingar i linje med sina mål. En viktig del av denna förmåga är situationsmedvetenhet, som innebär att förstå miljöns nuvarande tillstånd och förutse dess framtida utveckling. Traditionella autonoma system hanterar perception och prediktion som separata steg i en sekventiell kedja, där sensordata bearbetas till alltmer abstrakta representationer. Även om detta strukturerade tillvägagångssätt lett till stora framsteg, begränsas det av sensorbrister, inklusive skymda objekt, mätosäkerhet och ogynnsamt väder.

Denna avhandling undersöker hur prediktioner från tidigare observationer kan förbättra perceptionsalgoritmer, så att agenter kan sluta sig till saknad information, minska osäkerhet och bättre förutse förändringar. För att möjliggöra denna integration utforskas alternativa omgivningsrepresentationer som ger återkoppling mellan prediktion och perception, samtidigt som osäkerheter kan hanteras. Denna tätare koppling förbättrar beslutsfattandet, särskilt i komplexa och delvis observerbara miljöer.

Avhandlingens huvudsakliga bidrag inkluderar: (1) ett reso\-nemangs\-ramverk baserat på nåbarhet för att spåra möjliga dolda hinder; (2) dess utvidgning för att hantera fördröjd och ofullständig extern data; (3) en probabilistisk kartmetod, Transitional Grid Maps (TGM), som gemensamt modellerar statisk och dynamisk ockupation; och, (4) utvidgning av TGM för att förbättrad hantering av väderrelaterat sensorbrus.

Metoderna utvärderas i scenarier där traditionella perceptionskedjor har problem, exempelvis i skymda, mycket dynamiska och brusiga miljöer. Genom att överbrygga klyftan mellan perception och prediktion bidrar detta arbete till utvecklingen av robustare och intelligentare autonoma system.

Place, publisher, year, edition, pages
Stockholm: KTH Royal Institute of Technology, 2025. p. 71
Series
TRITA-ITM-AVL ; 2025:29
National Category
Robotics and automation
Research subject
Machine Design
Identifiers
urn:nbn:se:kth:diva-363919 (URN)978-91-8106-330-1 (ISBN)
Public defence
2025-06-18, https://kth-se.zoom.us/j/66710325262, Kollegiesalen, Brinellvägen 8, Stockholm, 10:00 (English)
Opponent
Supervisors
Funder
Vinnova
Available from: 2025-05-27 Created: 2025-05-27 Last updated: 2025-06-09Bibliographically approved

Open Access in DiVA

No full text in DiVA

Authority records

Gaspar Sánchez, José ManuelBruns, LeonardTumova, JanaJensfelt, PatricTörngren, Martin

Search in DiVA

By author/editor
Gaspar Sánchez, José ManuelBruns, LeonardTumova, JanaJensfelt, PatricTörngren, Martin
By organisation
Mechatronics and Embedded Control SystemsRobotics, Perception and Learning, RPL
Robotics and automation

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 135 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • 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