PlumX Metrics
Embed PlumX Metrics

Autonomous Navigation of Tracked Robot in Uneven Terrains

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), ISSN: 1611-3349, Vol: 14274 LNAI, Page: 74-84
2023
  • 1
    Citations
  • 0
    Usage
  • 1
    Captures
  • 0
    Mentions
  • 0
    Social Media
Metric Options:   Counts1 Year3 Year

Metrics Details

Conference Paper Description

Since microdosing of chemical, biological, radiological, and nuclear (CBRN) contaminants is enough to cause great damage to humans, operating robots are widely used to handle CBRN-related tasks. However, how to improve the automation capabilities of these robots in uneven environments, such as autonomous navigation, is still a huge challenge. Current navigation methods usually set the scene as flat pavement, without considering the situation that the land slope exceeds a certain threshold. In order to explore ways of autonomous navigation in uneven environments, a 3D path planning and navigation method for the tracked robot is proposed in this paper, respecting applicable traversability constraints in uneven terrains. Firstly, a 3D graph-based map is built according to the occupancy map of the uneven environment. A set of spatial points, ensuring collision-avoidance of the robot, is randomly sampled in free space, and a vertex map is generated based on these vertices for robot traversing. Then, regarding the robot’s climbing and obstacle crossing ability as constraints, a path planning algorithm is used to search for the best path based on the Dijkstra algorithm. Finally, a fusion SLAM method based on LiDAR, IMU and RGB-D camera is used to achieve real-time localization, and the pure pursuit algorithm is used for navigation. The simulation results show that the proposed method can provide a safe and effective 3D path for the tracked robot and enable the robot’s autonomous navigation in uneven environments.

Provide Feedback

Have ideas for a new metric? Would you like to see something else here?Let us know