Robots with aerial-ground locomotion and manipulation capabilities are being introduced as a promising solution to increase energy saving and manipulation accuracy for activities like inspection and maintenance in industrial environments. However, the effective operation of hybrid robots in these scenarios can become complex and higher levels of robot autonomy are still needed. This paper presents a novel motion planning method that exploits the hybrid locomotion capabilities of hybrid robots to compute safe and efficient paths with velocity references between inspection points in cluttered environments. The method, which is built over the basis of the RRT* algorithm, integrates the hybrid nature of these aerial-ground robots into the planning process to generate optimal global paths according to different indices (operation time, power consumption). Moreover, the motion planner also includes Dynamics Awareness for robust robot operation and collision avoidance in highly-cluttered environments. This consists of an extension of the planning algorithm to integrate the dynamic behaviour of the controlled robot during the expansion of the search tree. The method has been applied to the inspection of pipe arrays in industrial scenarios like oil and gas refineries. Validation results where the execution of hybrid paths generated by the motion planner has been evaluated through simulation under different conditions have demonstrated the clear advantages of this kind of hybrid planning.
Caballero et al. (Sun,) studied this question.