In response to the strong randomness

slow convergence speed

and poor path quality of the RRT* algorithm

a bidirectional heuristic adaptive search RRT* algorithm is proposed in this paper. Firstly

through the bidirectional search strategy

grow a random tree from the start to the end respectively

and search in the workspace to make the algorithm converge faster. Secondly

the idea of gravitational field in the artificial potential field is introduced to make the random tree grow towards the target position when generating new nodes

so as to reduce the strong randomness of random tree sampling. Finally

an adaptive step size strategy was adopted to improve the algorithm′s obstacle avoidance ability and path smoothness. Experimental results have shown that the comprehensive effect of these improvement strategies enables the improved RRT* algorithm to quickly and effectively complete path planning tasks.