Motion control of a caterpillar robot using optimized feedback linearization and sliding mode controllers

Caterpillar robots are widely used in various industrial applications, including pipe inspection, access to dangerous locations in factories, passing through narrow holes during natural disasters, and even in processes such as endoscopy and colonoscopy. In this paper, two control schemes, namely optimized feedback linearization controller and optimized sliding mode controller are designed and compared for forward-motion control of a bio-inspired caterpillar robot, which contains five bars connected via torque joints. At first, the governing nonlinear dynamic equations of motion are presented. Then, feedback linearization controller and sliding mode controller are designed to deliver angular positions of the robot to desired set points. To compare the performance of the mentioned controllers, the related parameters are optimized. In this regard, an appropriate objective function is defined to minimize the stabilization errors and control inputs, simultaneously. It is noted that the optimization process is performed using a genetic algorithm. Finally, the performance of the optimized controllers in the presence of parameter uncertainties, sensor noise, and actuator disturbances is compared quantitatively in the viewpoints of the stabilization error and control effort.