Event-Triggered Hierarchical Planner for Autonomous Navigation in Unknown Environment
End-to-end deep neural network (DNN)-based motion planners have shown great potential in high-speed autonomous UAV flight. Yet, most existing methods only employ a single high-capacity DNN, which typically lacks generalization ability and suffers from high sample complexity. We propose a novel event...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-11-01
|
| Series: | Drones |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2504-446X/7/12/690 |
| _version_ | 1797381394326880256 |
|---|---|
| author | Changhao Chen Bifeng Song Qiang Fu Dong Xue Lei He |
| author_facet | Changhao Chen Bifeng Song Qiang Fu Dong Xue Lei He |
| author_sort | Changhao Chen |
| collection | DOAJ |
| description | End-to-end deep neural network (DNN)-based motion planners have shown great potential in high-speed autonomous UAV flight. Yet, most existing methods only employ a single high-capacity DNN, which typically lacks generalization ability and suffers from high sample complexity. We propose a novel event-triggered hierarchical planner (ETHP), which exploits the bi-level optimization nature of the navigation task to achieve both efficient training and improved optimality. Specifically, we learn a depth-image-based end-to-end motion planner in a hierarchical reinforcement learning framework, where the high-level DNN is a reactive collision avoidance rerouter triggered by the clearance distance, and the low-level DNN is a goal-chaser that generates the heading and velocity references in real time. Our training considers the field-of-view constraint and explores the bi-level structural flexibility to promote the spatio–temporal optimality of planning. Moreover, we design simple yet effective rules to collect hindsight experience replay buffers, yielding more high-quality samples and faster convergence. The experiments show that, compared with a single-DNN baseline planner, ETHP significantly improves the success rate and generalizes better to the unseen environment. |
| first_indexed | 2024-03-08T20:50:52Z |
| format | Article |
| id | doaj.art-8a1376671bb048ea92fa332d61f24e82 |
| institution | Directory Open Access Journal |
| issn | 2504-446X |
| language | English |
| last_indexed | 2024-03-08T20:50:52Z |
| publishDate | 2023-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Drones |
| spelling | doaj.art-8a1376671bb048ea92fa332d61f24e822023-12-22T14:03:57ZengMDPI AGDrones2504-446X2023-11-0171269010.3390/drones7120690Event-Triggered Hierarchical Planner for Autonomous Navigation in Unknown EnvironmentChanghao Chen0Bifeng Song1Qiang Fu2Dong Xue3Lei He4School of Aeronautics, Northwestern Polytechnical University, Xi’an 710072, ChinaSchool of Aeronautics, Northwestern Polytechnical University, Xi’an 710072, ChinaSchool of Aeronautics, Northwestern Polytechnical University, Xi’an 710072, ChinaSchool of Aeronautics, Northwestern Polytechnical University, Xi’an 710072, ChinaSchool of Aeronautics, Northwestern Polytechnical University, Xi’an 710072, ChinaEnd-to-end deep neural network (DNN)-based motion planners have shown great potential in high-speed autonomous UAV flight. Yet, most existing methods only employ a single high-capacity DNN, which typically lacks generalization ability and suffers from high sample complexity. We propose a novel event-triggered hierarchical planner (ETHP), which exploits the bi-level optimization nature of the navigation task to achieve both efficient training and improved optimality. Specifically, we learn a depth-image-based end-to-end motion planner in a hierarchical reinforcement learning framework, where the high-level DNN is a reactive collision avoidance rerouter triggered by the clearance distance, and the low-level DNN is a goal-chaser that generates the heading and velocity references in real time. Our training considers the field-of-view constraint and explores the bi-level structural flexibility to promote the spatio–temporal optimality of planning. Moreover, we design simple yet effective rules to collect hindsight experience replay buffers, yielding more high-quality samples and faster convergence. The experiments show that, compared with a single-DNN baseline planner, ETHP significantly improves the success rate and generalizes better to the unseen environment.https://www.mdpi.com/2504-446X/7/12/690autonomous UAV flightevent-triggeredhierarchical reinforcement learningcollision avoidanceunknown environment |
| spellingShingle | Changhao Chen Bifeng Song Qiang Fu Dong Xue Lei He Event-Triggered Hierarchical Planner for Autonomous Navigation in Unknown Environment Drones autonomous UAV flight event-triggered hierarchical reinforcement learning collision avoidance unknown environment |
| title | Event-Triggered Hierarchical Planner for Autonomous Navigation in Unknown Environment |
| title_full | Event-Triggered Hierarchical Planner for Autonomous Navigation in Unknown Environment |
| title_fullStr | Event-Triggered Hierarchical Planner for Autonomous Navigation in Unknown Environment |
| title_full_unstemmed | Event-Triggered Hierarchical Planner for Autonomous Navigation in Unknown Environment |
| title_short | Event-Triggered Hierarchical Planner for Autonomous Navigation in Unknown Environment |
| title_sort | event triggered hierarchical planner for autonomous navigation in unknown environment |
| topic | autonomous UAV flight event-triggered hierarchical reinforcement learning collision avoidance unknown environment |
| url | https://www.mdpi.com/2504-446X/7/12/690 |
| work_keys_str_mv | AT changhaochen eventtriggeredhierarchicalplannerforautonomousnavigationinunknownenvironment AT bifengsong eventtriggeredhierarchicalplannerforautonomousnavigationinunknownenvironment AT qiangfu eventtriggeredhierarchicalplannerforautonomousnavigationinunknownenvironment AT dongxue eventtriggeredhierarchicalplannerforautonomousnavigationinunknownenvironment AT leihe eventtriggeredhierarchicalplannerforautonomousnavigationinunknownenvironment |