An Improved RRT* Path Planning Algorithm for Service Robot

Abstract

The space environment of path planning for service robots is complex. In this environment, the traditional Rapid-exploring Random Tree * (RRT*) algorithm guarantees the completeness and asymptotic optimality of global search probability, but it has high memory occupancy, slow convergence speed, large sampling space and long time-consuming. Therefore, a fast extended random tree star trajectory planning algorithm is proposed in this paper. R-RRT* improves the above problems through three novel strategies: connected domain sampling, elliptic modeling sampling and path optimization strategy. Two kinds of complex map scenarios are simulated and compared. The simulation results show that the method can locate the target area quickly and improve the computational efficiency. Compared with the traditional RRT* method, it has the advantages of low memory occupancy, fast convergence speed and short time-consuming.