Research On The Free Gait Planning And Motion Mechanism Of A Hexapod Robot

Due to the abundant gait and redundant structure, the hexapod robot can take advantage of discrete support points to achieve obstacle avoidance, obstacle crossing, and free walking on uneven terrain and has better adaptability for complex terrains compared with wheeled and tracked mobile robots. With the development of basic theories and key technologies of hexapod robots, it can be expected that the hexapod robot will be applied in a wide range of areas. This work is supported by the National High Technology Research and Development Program of China(2012AA041508), the Natural Science Foundation of Hebei Province(E2014202154), and the State Key Laboratory of Robotics and System(SKLRS-2013-ZD-04). In order to improve performance of hexapod robots in the complex nonstructural environment, the basic theories and key technologies related to the hexapod robot are researched. Innovative work of this paper includes:1. Based on the observation experiments, generic structures of six-legged creatures are researched and the structure of the hexapod robot is designed. Considering the overall performance, the key structure parameters are optimized. Then the structural performance is verified by virtual prototyping technology. On this basis, a semi-autonomous control system is designed for the hexapod robot and a new hexapod robot is developed.2. For multi-legged coordinated control, an efficient parametric foot trajectory planning method is proposed. The hexapod robot is regarded as a time-varying parallel mechanism and the determination methods and boundary equations of its workspace are described. Based on the relativity of motions, an efficient body motion planning method is presented. On this basis, a new multi-legged coordinated control method is proposed, and different simulations are conducted to verify the effectiveness of the method.3. For the free gait planning, a discrete model of stepping is built. Based on stability analysis, a stable state space is constructed and the issue of gait planning is transformed into the reordering problem of states in stable state space. Then new free gait generation algorithms are proposed.4. Considering specific working conditions, optimized free gait generation methods for structured and nonstructured environment are presented respectively based on average stability margin. Based on the Markov decision processes, free gait generation algorithm based on reinforcement learning is proposed. All the algorithms are verified by simulations.5. An experiment platform for the hexapod robot is built and some experiments related to multi-legged coordinated control and free gait planning are carried out to verify the rationality of the structure and the accuracy of the theoretical analysis.