Research on path planning method for staircase turning platforms

For indoor path planning, it is essential to account for inter-floor navigation requirements. The unmanned vehicle must navigate the turning platform when ascending or descending stairs. These narrow platforms, combined with the vehicle's large size, increase the risk of collisions and scrapes while turning. To address this issue, an improved A* algorithm is proposed. This method first expands obstacles when generating a 2D grid map and improves the cost function of the A* algorithm by introducing the reciprocal distance between node n and the handrail of the turning platform, to avoid obstacles as well as adding a turning cost function to make the planned path smoother. By comprehensively considering these costs, a safer and more feasible path can be planned. Next, the ‘arc collision detection method’ is used to calculate whether the arc of the vehicle's diagonal during a turn overlapped with surrounding obstacles, effectively preventing the vehicle from colliding or scraping against obstacles while turning. Then, Bezier curve is used to smooth the path, solving the problem of sharp turns that are unfavorable for vehicle movement. Simulation experiments demonstrate that, with this method, the unmanned vehicle can safely navigate stair turning platforms, achieving shorter path lengths and planning times compared to traditional algorithms..