论文部分内容阅读
在非结构化的空间中,现有的移动机械人局部路径规划所获得的信息是有限的,不利于机器人的局部路径规划;为了解决这个问题,提出了一种多传感器信息融合的局部路径规划,并推导了一种基于改进的BP神经网络-模糊逻辑法的局部路径规划算法;利用BP神经网络在学习、分类和优化上的优势,采集的非线性信息首先经过BP神经网络学习推理,进行融合;再利用模糊逻辑法进行信息处理,在初速度V=0.5m/s,时间t在区间[0,19.9]上,采样频率为5Hz的环境中,进行正弦轨迹的局部路径规划,算法输出结果的误差在-0.025～0.025之间;该算法改善了移动机器人的局部路径规划的效果。
In unstructured space, the information obtained by the existing mobile robot local path planning is limited, which is not good for the robot’s local path planning. In order to solve this problem, a local path planning based on multi-sensor information fusion , And a local path planning algorithm based on improved BP neural network-fuzzy logic is deduced. Based on the advantages of BP neural network in learning, classification and optimization, the collected nonlinear information is first learned through BP neural network reasoning Then the fuzzy logic method is used to process the information. Local path planning of the sine locus is performed in the environment where the initial velocity V = 0.5 m / s, the time t in the interval [0, 19.9] and the sampling frequency 5 Hz, and the algorithm outputs The error of the result is between -0.025 ~ 0.025; this algorithm improves the effect of local path planning of mobile robot.