Autonomous driving using imitation learning with look ahead point for semi structured environments
Abstract Semi-structured environments are difficult for autonomous driving because there are numerous unknown obstacles in drivable area without lanes, and its width and curvature considerably change. In such environments, searching for a path on a real-time is difficult, and localization data are i...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Nature Portfolio
2022-12-01
|
Series: | Scientific Reports |
Online Access: | https://doi.org/10.1038/s41598-022-23546-6 |
_version_ | 1811302371214491648 |
---|---|
author | Joonwoo Ahn Minsoo Kim Jaeheung Park |
author_facet | Joonwoo Ahn Minsoo Kim Jaeheung Park |
author_sort | Joonwoo Ahn |
collection | DOAJ |
description | Abstract Semi-structured environments are difficult for autonomous driving because there are numerous unknown obstacles in drivable area without lanes, and its width and curvature considerably change. In such environments, searching for a path on a real-time is difficult, and localization data are inaccurate, reducing path tracking accuracy. Instead, alternative methods that reactively avoid obstacles in real-time using candidate paths or an artificial potential field have been studied. However, these require heuristics to identify specific parameters for handling various environments and are vulnerable to inaccurate input data. To address these limitations, this study proposes a method in which a vehicle drives toward drivable area using vision and deep learning. The proposed imitation learning method learns the look-ahead point where the vehicle should reach on a vision-based occupancy grid map to obtain a safe policy with a clear state action pattern relationship. Furthermore, using this point, the data aggregation (DAgger) algorithm with weighted loss function is proposed, which imitates expert behavior more accurately, especially in unsafe or near-collision situations. Experimental results in actual semi-structured environments demonstrated the limitations of general model-based methods and the effectiveness of the proposed imitation learning method. Moreover, simulation experiments showed that DAgger with the weight obtains a safer policy than existing DAgger algorithms. |
first_indexed | 2024-04-13T07:26:58Z |
format | Article |
id | doaj.art-64b484c319e144a4a8b4d28abf37d91a |
institution | Directory Open Access Journal |
issn | 2045-2322 |
language | English |
last_indexed | 2024-04-13T07:26:58Z |
publishDate | 2022-12-01 |
publisher | Nature Portfolio |
record_format | Article |
series | Scientific Reports |
spelling | doaj.art-64b484c319e144a4a8b4d28abf37d91a2022-12-22T02:56:27ZengNature PortfolioScientific Reports2045-23222022-12-0112111710.1038/s41598-022-23546-6Autonomous driving using imitation learning with look ahead point for semi structured environmentsJoonwoo Ahn0Minsoo Kim1Jaeheung Park2Dynamic Robotic Systems (DYROS) Lab., Graduate School of Convergence Science and Technology, Seoul National UniversityDynamic Robotic Systems (DYROS) Lab., Graduate School of Convergence Science and Technology, Seoul National UniversityDynamic Robotic Systems (DYROS) Lab., Graduate School of Convergence Science and Technology, Seoul National UniversityAbstract Semi-structured environments are difficult for autonomous driving because there are numerous unknown obstacles in drivable area without lanes, and its width and curvature considerably change. In such environments, searching for a path on a real-time is difficult, and localization data are inaccurate, reducing path tracking accuracy. Instead, alternative methods that reactively avoid obstacles in real-time using candidate paths or an artificial potential field have been studied. However, these require heuristics to identify specific parameters for handling various environments and are vulnerable to inaccurate input data. To address these limitations, this study proposes a method in which a vehicle drives toward drivable area using vision and deep learning. The proposed imitation learning method learns the look-ahead point where the vehicle should reach on a vision-based occupancy grid map to obtain a safe policy with a clear state action pattern relationship. Furthermore, using this point, the data aggregation (DAgger) algorithm with weighted loss function is proposed, which imitates expert behavior more accurately, especially in unsafe or near-collision situations. Experimental results in actual semi-structured environments demonstrated the limitations of general model-based methods and the effectiveness of the proposed imitation learning method. Moreover, simulation experiments showed that DAgger with the weight obtains a safer policy than existing DAgger algorithms.https://doi.org/10.1038/s41598-022-23546-6 |
spellingShingle | Joonwoo Ahn Minsoo Kim Jaeheung Park Autonomous driving using imitation learning with look ahead point for semi structured environments Scientific Reports |
title | Autonomous driving using imitation learning with look ahead point for semi structured environments |
title_full | Autonomous driving using imitation learning with look ahead point for semi structured environments |
title_fullStr | Autonomous driving using imitation learning with look ahead point for semi structured environments |
title_full_unstemmed | Autonomous driving using imitation learning with look ahead point for semi structured environments |
title_short | Autonomous driving using imitation learning with look ahead point for semi structured environments |
title_sort | autonomous driving using imitation learning with look ahead point for semi structured environments |
url | https://doi.org/10.1038/s41598-022-23546-6 |
work_keys_str_mv | AT joonwooahn autonomousdrivingusingimitationlearningwithlookaheadpointforsemistructuredenvironments AT minsookim autonomousdrivingusingimitationlearningwithlookaheadpointforsemistructuredenvironments AT jaeheungpark autonomousdrivingusingimitationlearningwithlookaheadpointforsemistructuredenvironments |