Motion control of obstacle avoidance for the robot arm via improved path planning algorithm

To improve the shortcomings of the traditional RRT (Rapid exploration of random tree) algorithm, the G-RRT is introduced as an enhanced robotic arm path planning method. The G-RRT can be also used to improve search efficiency and reduce computational complexity through the introduction of probability thresholds, dynamic step expansion and simplified random trees. The G-RRT is compared with conventional RRT and P-RRT (RRT with target bias strategy only) through simulation experiments. The results demonstrate the effectiveness of G-RRT and show optimized path planning. Furthermore, obstacle avoidance tests on conventional RRT and G-RRT algorithms show significant improvements in the latter. The G-RRT algorithm generates a 99% reduction in the number of random tree nodes, a 97.8% reduction in search time, and a 32% reduction in path length compared to the traditional RRT algorithm. The results show that the G-RRT algorithm takes less time to plan and the path is shorter. These improvements provide a reference for the application of intelligent technology in actual scenarios.