Skid-steered mobile robots are often used in outdoor exploration due to their robust mechanical structure and high maneuverability. When they track reference path on a slope with boundaries, ensuring the tracking accuracy and stability of the skid-steered mobile robot is the major target. However, the gravity makes the relationship between wheels and ground more complex on the slope, and variational slope angle also makes it difficult for tracking control. The common control methods focus on plane motion, where only the plane forces are taken into account and the gravity is normally ignored. It may lead to some performance limitations such as the accuracy of motion on a slope. To address these problems, a model predictive control strategy combined with a fuzzy system is proposed in this paper, which has considered the dynamics of the body and wheels on the slope. We improved the two dimensional kinematics and dynamics model of the robot, which makes the three dimensional motion control more accurate. And the control method allows the robot to adapt to slopes with different angles and to make the path tracking stable to curvature mutation. Both experiment and simulation results demonstrate the effectiveness and superiority of the proposed model and method.
Skid-steered mobile robots are often used in outdoor exploration due to their robust mechanical structure and high maneuverability. When they track reference path on a slope with boundaries, ensuring the tracking accuracy and stability of the skid-steered mobile robot is the major target. However, the gravity makes the relationship between wheels and ground more complex on the slope, and variational slope angle also makes it difficult for tracking control. The common control methods focus on plane motion, where only the plane forces are taken into account and the gravity is normally ignored. It may lead to some performance limitations such as the accuracy of motion on a slope. To address these problems, a model predictive control strategy combined with a fuzzy system is proposed in this paper, which has considered the dynamics of the body and wheels on the slope. We improved the two dimensional kinematics and dynamics model of the robot, which makes the three dimensional motion control more accurate. And the control method allows the robot to adapt to slopes with different angles and to make the path tracking stable to curvature mutation. Both experiment and simulation results demonstrate the effectiveness and superiority of the proposed model and method.