一种基于假设检验的机器人所处区域类型识别方法

为实现机器人对其所处区域的有效识别,提出一种基于假设检验的区域类型识别方法。首先考虑观测误差影响提出一种基于概率的未知障碍物识别方法。进而将观测信息视为对周围环境的采样,假设机器人所处区域类型,利用观测信息中的未知障碍物数对其验证,实现对区域类型的识别。该方法考虑了实际中观测误差的影响,限制了误判的概率。实验证明,该方法能够在观测误差影响下有效识别机器人所处区域类型,并成功将其应用于部分未知环境的路径规划中。

Recognition approach of regional type of mobile robotbased on hypothesis testing theory

This paper proposed a recognition approach of regional type based on hypothesis testing theory. It firstly proposed a probabilistic identification method of unexpected obstacle considering the uncertainty of sensor information. Then it treated sensor information as a sampling of the environment around the robot, so that it tested a hypothesis of the regional type of the environment confronting the robot, using the number of the sensor measurement caused by unexpected obstacle. This approach successfully limited the probability of misjudgement. Experiment result proves that the proposed approach is able to recognize the regional type considering the inaccurate sensing information. More than that, this paper proposed a path planning experiment under partly unknown environment using the proposed approach to demonstrates its effect.