移动机器人RRT算法改进研究

目前,机器人路径规划常用算法有避障(Bug)算法、概率路线图(PRM)算法、快速搜索随机树(RRT)算法、蚁群算法、人工势场法等,其中RRT算法在路径规划中应用最广。针对RRT算法存在随机性强、偏差大、路径不一定最优、收敛速度慢等缺点,对RRT算法进行改进,引导随机树向目标点生长,借助人工势场的引力思想,并加入自适应策略,通过机器人与目标点位置、速度和加速度的不断变化来改变步长大小,使机器人快速到达目标点。实验结果表明,通过自适应RRT算法可以提高算法收敛性,缩短了算法时间,可以有效应用在移动机器人系统上,提高移动机器人的工作效率。

Improved RRT Algorithm for Mobile Robot

At present,the commonly used algorithms for robot path planning include Bug algorithm,probabilistic road maps(PRM) algorithm,rapidly-exploring random trees(RRT) algorithm,ant colony algorithm,antificial potential field method,among which RRT algorithm is the most widely used in path planning.In order to solve the shortcomings of RRT algorithm,such as strong randomness,large deviation,not necessarily optimal path and slow convergence speed,the RRT algorithm is improved to guide the random tree to grow to the target point.With the help of the gravity idea of artificial potential field and the addition of adaptive strategy,the step size is changed by the constant changes of the position,speed and acceleration of the robot and the target point,which can make the robot reach the target point quickly.The experimental results show that the adaptive RRT algorithm can improve the convergence of the algorithm,shorten the algorithm time,and can be effectively applied to the mobile robot system to improve the work efficiency of the mobile robot.