For the problem of robot path planning, a geometry intersection testing based path planning algorithm is presented.
Firstly, the first obstacle grid is searched in the configuration space which lies on the oriented segment from start to target. By
combining the greedy method and backward strategy, the grid where next way point lies is determined from the neighbors of
the first obstacle grid by local searching. Then subsequent way points are obtained iteratively scratching from the last obtained
way point in order to get some routes from start to target as possible. Finally, the best route is refined by incorporating sub-
routes to maintain the consistency of trajectory. The experimental results show that, the presented method can plan a shorter
and more reasonable path than other algorithms, which represents higher consistency of trajectory, and the computing time
can meet the requirement of practical application.