论文部分内容阅读
结合栅格法与神经元基本思想,提出了一种新的移动机器人完全遍历路径规划方法。该方法首先利用栅格法对环境地图进行划分,把划分而得到的每一个栅格作为一个神经元,然后将神经元进行初始活性赋值,依据机器人运动方向对神经元实施外部激励;机器人通过寻求邻近神经元中活性值最大的神经元而移动,直至遍历完环境地图上所有可达点。计算机仿真实验表明:该方法能有效避障、路径重复率小,实现了机器人的完全遍历路径规划,也验证了该方法的可行性与有效性。
Combining the basic idea of raster and neuron, a new path planning method for fully traversed mobile robot is proposed. In this method, firstly, the grid is used to divide the environment map, each grid is divided into a neuron, and then the neurons are assigned the initial activity, and the neurons are externally excited according to the movement direction of the robot; Move near the neurons with the highest activity value in neurons until all the reachable points on the environment map have been traversed. Computer simulation results show that this method can effectively avoid obstacles, and the path repetition rate is small, and the complete traversal path planning of the robot is realized, and the feasibility and effectiveness of the method are also verified.