Safety Assessment of Trajectories for Navigation in Uncertain and Dynamic Environments

摘要：

This paper presents a probabilistic threat assessment method for reasoning about the safety of robot trajectories in uncertain and dynamic environments. For safety evaluation, the overall collision probability is used to rank candidate trajectories by considering the collision probability of known objects as well as the collision probability beyond the planning horizon. Monte Carlo sampling is used to estimate the collision probabilities. This concept is applied to a navigation framework that generates and selects trajectories in order to reach the goal location while minimizing the collision probability. Simulation scenarios are used to validate the overall crash probability and show its necessity in the proposed navigation approach.

作者: Daniel Althoff Dirk Wollherr Martin Buss

作者单位: Institute of Automatic Control Engineering (LSR)

会议类型: 国际会议

会议名称: 2011 IEEE International Conference on Robotics and Automation(2011年IEEE世界机器人与自动化大会 ICRA 2011)

会议地点: 上海

会议语种:英文

页码: 5407-5412

在线出版日期: 2011-05-09（万方平台首次上网日期，不代表论文的发表时间）