Control System Design And Obstacle-Climbing Analysis For A Leg-Wheeled Mobile Robot

In modern wars, every country is trying to achieve―zero mortality‖. The casualty is required to be as low as possible with the expected martial purpose achieved. Due to such military need, unmanned troops, including unmanned aerial vehicle and unattended ground weapon platform, will become an importance force in future warfare. This paper completes the following research with the laboratory project―leg-wheeled robot‖.Firstly, the design, assembly and debugging of the leg-wheeled robot is completed. The robot uses six sets of wheel-leg device; the movement of the legs and wheels are separately controlled by leg motor and motion motor. The control system is divided into two parts: the embedded computer control system and the motion controller. The embedded computer control system achieves the robot's planar motion and obstacle negotiation algorithm, produces the direction of the motor's movement, such as speed and location. The motion controller achieves the motion control of the motor, including speed PID and location PID. The communication between these two system adopts CAN bus.Secondly, the leg-wheeled robot is kinetically modeled and analyzed. The kinemics of the robot is divided into two parts: the kinemics of planar motion and the kinetics of spatial posture. In the kinemics of planar motion, the differential steering of the robot's planar motion is modeled kinetically; the robot's speed distribution is analyzed through the design of motion tracks. In spatial kinetics, the robot's general spatial kinetic model and CG kinetic model are established; according to the special posture of the robot during the obstacle negotiation, the general CG kinetic model is simplified. The CG kinetic model in different special posture is obtained, which provides reference for the stability control of the robot's obstacle negotiation.Thirdly, the leg-wheeled robot's ability to overcome typical obstacles is analyzed. For vertical obstacle, ditch obstacle and slope obstacle, the robot's negotiation ability is analyzed, the negotiation action is planned; the optimization method incorporating obstacle margin, safety and stability is proposed, and the obstacle posture is optimized numerically. The planned obstacle movement is virtually simulated in ADAMS, which proves the validity of the obstacle-clearing movement.Last, the human-computer interface to control the robot is designed. The experiment on the robot's planar kinetics and obstacle-clearing is completed, and the correctness of the control theory proposed in this article is verified.