Control for a Two-Link Planar Robot With an Actuated Tail

In this paper, a new climbing robot concept and its control strategy are introduced. The robot presented in this article is a two-link planar robot equipped with two magnetic grippers and one actuated tail, which is designed to add energy to the system for efficient climbing. The climbing strategy for this robot is inspired by brachiation robots. By alternating the activation of the grippers and using the tail to add energy, the robot can string together a sequence of pendulum-like paths with brachiation motion to reach any configuration on the climbing surface. Its high mobility and low energy consumption make this robot a candidate for applications in cleaning and inspection on vertical surfaces. The non-linear dynamics of the swinging motion for this robot are studied using the Lagrange formulation under the assumption that there is no slip on the grippers. The control strategy of this robot uses a high-level planner to determine the grip points and a low-level controller to actuate the tail, achieving low energy consumption with high accuracy. Model predictive control (MPC) is applied for the low-level swinging motion controller to determine the optimal torque provided to the tail. In this initial work the robot trajectory is planning based on minimization of the distance travelled. The performance of this robot is discussed based on MATLAB simulation results. The paper concludes by discussing alternative planning/control strategies along with initial work towards experimental validation.