Design and control methods for a hexapod flying robot in complex terrain

This paper presents a hexapod aerial robot and its control method designed for complex terrain. The robot consists of a main body with legs distributed around it, each equipped with propeller assemblies for transitioning between flight and walking modes. The control method involves graying the images captured by a camera, extracting texture features from the grayed images[1], constructing a gray-level co-occurrence matrix, and classifying and identifying obstacles. Based on the required obstacle avoidance mode[2], the robot determines whether it is in a walking or flying posture. When in walking mode, the motion parameters for the robot’s legs are constructed, and when in flying mode, the power allocation parameters for the robot’s rotors are determined. This approach, which does not require replacing any robot components, significantly enhances the robot’s environmental adaptability[3], improves stability and mobility, optimizes the utilization of each component, leads to a more compact structure, and increases both space utilization and overall adaptability.