This letter proposes the Polygon-based Random Tree search algorithm with Variable polygon size (PRT-V) for Variable Topology Truss (VTT). This algorithm improves on our previous path planning algorithm by using VTT’s ability to change its size drastically. The algorithm partitions the space into unconstrained areas where normal (higher speed) locomotion can work using our previous technique and tighter space areas where the robot will need to shrink in size. We call the points that delimit these spaces stopover points. A Voronoi diagram-based roadmap method is used to automatically find appropriate stopover points. Based on the stopover points, the desired support polygons and the corresponding nominal lengths are generated by PRT-V. Then, the locomotion of VTT is generated by a non-impact locomotion algorithm to prevent tipping over. Simulations verify the performance of PRT-V in three different workspaces with obstacles. With PRT-V, VTT can squeeze through narrow passages or move efficiently in large spaces by increasing size, a unique feature.