论文部分内容阅读
为满足自主导航轨迹定性和定量评测要求,提出一种基于不确定性云模型的轨迹分析计算方法.该方法通过提取机器人自主行驶和避障行为中的运动轨迹特征,建立对应的方位偏离、方向偏差和避障安全距离轨迹特征云模型,以云模型的期望作为轨迹特征基本度量,以熵和超熵表达特征所具有的模糊性和随机性.利用云模型所具有的不确定性度量优势,反映系统自主导航过程中的瞬时状态和系统稳定性.实验结果表明,该方法能够有效评估自主导航系统轨迹数据,并能够弥补基于增强学习的方法无法直观反映系统评价稳定性的不足.
In order to meet the requirements of qualitative and quantitative evaluation of autonomous navigation trajectory, a trajectory analysis and calculation method based on the uncertainty cloud model is proposed in this paper. By extracting the characteristics of motion trajectory of autonomous navigation and obstacle avoidance behavior of robot, the corresponding azimuth deviation, Deviation and obstacle avoidance safety distance trajectory feature cloud model, the expectation of cloud model is taken as the basic measure of trajectory feature, and the entropy and hyperspensity expression features have the fuzziness and randomness.Using the advantage of cloud model’s uncertainty measurement, Which reflects the instantaneous state and system stability in the process of autonomous navigation.The experimental results show that this method can effectively evaluate the trajectory data of autonomous navigation system and can make up for the deficiency that the method based on reinforcement learning can not directly reflect the stability of system evaluation.