Terrain Adaptive Gait Transitioning for a Quadruped Robot using Model Predictive Control

Abstract

Legged robots can traverse challenging terrain, use perception to plan their safe foothold positions, and navigate the environment. Such unique mobility capabilities make these platforms a perfect candidate for scenarios such as search and rescue, inspection, and exploration tasks. While traversing through such terrains, the robot's instability is a significant concern. Many times the robot needs to switch gaits depending on its environment. Due to the complex dynamics of quadruped robots, classical PID control fails to provide high stability. Thus, there is a need for advanced control methods like the Model Predictive Control (MPC) which uses the system model and the nature of the terrain in order to predict the stable body pose of the robot. The controller also provides correction to any external disturbances that result in a change in the desired behavior of the robot. The MPC controller is designed in MATLAB, for full body torque control. The controller performance was verified on Boston Dynamics Spot in Webots simulator. The robot is able to provide correction for external perturbations up to 150 N and also resist falls till 80 cm.

Reduced Model Predictive Control Toward Highly Dynamic Quadruped Locomotion

“…Capitalizing on this advantage, researchers have introduced convex MPC based on SRBM with a predefined foot trajectory [37]. This approach has been employed to simulate or experiment with dynamic motions, including gallop gait [38], back-flips [39], and gait transitions [40]. To address problems associated with Euler angles and quaternion representation for orientation, a variation-based modeling method was proposed [41], and its effectiveness was demonstrated through implementations of wall-climbing [42], back-flips [43] and attitude control [44].…”

Development of a 3D Printed Biologically Inspired Monoped Self-Balancing Robot

“…The jump control consists of modelling the robot around essential parameters like angle of attack, take-off direction, velocity and stability, righting reflex, landing buffering and self-correction mechanisms [6]. As a reference, at the writing time of this paper, there are a few robotics developments that have successfully reached a high level of complexity and functionality and pushed the study of jumping locomotion a bit farther: Spot from Boston Dynamics [7], Salto 1P from UC Berkeley [8], Leonardo from Caltech [9], Ascento 2 from ETH Zurich [10] and Hugh-Flying Jumper from University of California [11] (Fig. 3).…”