Measuring the Influence of Environmental Factors on Obstacle Detection and Avoidance with an Autonomous Ground Vehicle
Abstract
This paper presents the results of a comprehensive study of obstacle detection and avoidance(ODOA) by an autonomous ground vehicle (AGV) in off-road and adverse environmental conditions. This study included both real and simulated testing of an AGV operating in challenging conditions such as rain, dust, and deformable terrain. A novel approach for analyzing the environmental impact on each subsystem (perception, planning, control) of the vehicle was implemented in simulation and used to evaluate multiple options for planning and perception algorithms. This work is the most complete and systematic test campaign of its kind to be conducted on a publicly available autonomy stack and will facilitate the development of test strategies for AGV in future work. The primary contributions of this work are the development of a free and open source autonomous software stack for off-road AGV, a method for quantitative assessment of AGV systems, and incorporation of combined simulated and physical testing into a comprehensive test approach. This work demonstrates how simulation can be used to measure aspects of AGV performance that are impossible to measure in physical tests, giving additional insight into the functioning of the autonomy stack.
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 04, 2022
- Accession Number
- AD1177696
Entities
People
- Christopher Goodin
- Christopher H. Hudson
- Daniel W. Carruth
- Lalitha Dabbiru
- Lucas D. Cagle
- Marc N. Moore
- Nicklaus Scherrer
- Paramsothy Jayakumar
Organizations
- Mississippi State University
- United States Army Tank Automotive Research, Development and Engineering Center