Motion And Force Planning For Hexapod Robots Traversing Uneven Terrains Based On Force Constraints

Planet exploration,disaster rescue,and transportation in the mountains have brought many challenges to mobile robots.These complex environments are full of gullies,rocks,and ruins so that traditional wheeled and tracked robots cannot traverse.The hexapod robots have the advantages of good stability,strong load capacity,and strong terrain adaptability.They have a concrete ability to pass through rough terrains.Due to the complexity of rough terrains and the overconstrained characteristics of the hexapod robots,Its feet have position deviations in both normal and tangential directions,resulting in internal forces of all supporting legs.The motion planning and force distribution of hexapod robots that equilibrium constraints are considered in rough terrains have sufficiently been explored in this dissertation,aiming to improve the traversing ability in uneven terrains.The foothold sequence planning method that satisfies the quasi-static equilibrium constraints is proposed in this dissertation.The stability of hexapod robots in rough terrains is analyzed,and constraints that guarantee the hexapod robots’ stability are modeled,including force and torque balance constraint,normal contact force constraint,joint torque constraint,and friction cone constraint.The 3D quasi-static equilibrium support region(3D QESR)is proposed to guarantee the stability of single-step motion.The foothold sequence planning method based on the heuristic search is proposed,and the 3D QESR is one of the search method’s constraints to guarantee the stability of planned results.Several rough terrains are embedded in Gazebo to verify the proposed method,and simulation results verified the method’s effectiveness.The foothold sequence is discrete,and the hexapod robots must have a continuous trajectory that connects adjacent footholds before gait walking.The trajectory of hexapods includes trunk trajectory,swing legs trajectory,and supporting legs trajectory.The trunk trajectory planning is transformed into a quadratic programming problem.3D QESR is the constraint to guarantee the stability of hexapod robots,and minimum jerk is the optimization objective of this QP problem.The virtual terrain field based on force disturbance is introduced to solve the unexpected collision of swing legs.The swing legs’ re-planning strategy is proposed,and the prototype experiments are used to verify the strategy’s effectiveness.In order to avoid the internal force of all supporting legs,the desired contact force and joint torque of hexapod robots should be planned.In order to prevent the feet from slipping due to uneven force distribution and enhance the climbing ability of hexapod robots,the decomposed quadratic programming contact force distribution(DQP-based CFD)method is proposed to reduce the traction coefficient.In order to reduce the energy consumption of the hexapod robot,its instantaneous power model is established and transformed into the objective function.The joint torque distribution that can reduce the energy consumption of the hexapod robot is proposed.The contact force and joint torque distribution method are compared with the existing methods by numerical simulation,which proves the effectiveness.A compliance control framework based on contact force distribution is built,and simulations and experiments results proved the feasibility and effectiveness of the method.In order to verify the motion planning and force distribution method of hexapod robots to traverse rough terrains,the software and hardware control architecture of the El Spider hexapod robot is built.The motion planning system of the hexapod robot is built by ROS(Robot Operating System),including the foothold sequence planning and trajectory planning.The force distribution and compliance control system is built by Twin CAT(The Windows Control and Automation Technology).An external sensorbased foothold sequence tracking strategy is proposed.This strategy adjusts each leg’s stride distance,forward direction,and foothold position during walking according to the error between the expected and actual poses.The prototype experiments are used to verify the effectiveness of the foothold sequence planning strategy.The motion planning experiments of the El Spider hexapod robot with compliant control is carried out,and a variety of indoor and outdoor rough terrains are used to verify the motion planning method and force distribution method comprehensively.