Balance of quadruped robots in off-road and soft soil environments

This research explores the walking gait of quadruped robots in off-road and soft soil environments, with a focus on addressing the challenge of foot entrapment during movement. Traditional crawler and wheeled robots face stability issues on uneven terrains, such as vineyards, where the rugged soil can cause overturning. Quadruped robots provide better adaptability due to their independent limb movements, but face the risk of sinking into soft soil, limiting their effectiveness in such conditions. In this study, we propose a control algorithm that detects when a leg becomes stuck in the soil and activates a struggle gait to extricate the robot. The system employs light sensors placed on the lower legs to detect when the robot’s foot begins to sink, triggering a change in gait to release the leg. We developed the core control algorithms for both the normal and the struggle gaits using MATLAB, with visualization achieved through MATLAB animations to represent the robot’s movements in different scenarios. The struggle gait algorithm ensures stable movement on challenging terrain, while the normal gait algorithm provides a solid baseline for standard conditions. This approach was demonstrated using MATLAB, integrating dynamic and kinematic models of the robot. The results demonstrate that the proposed solution offers a reliable way for quadruped robots to navigate soft soil environments without becoming trapped, improving their potential for agricultural and other off-road applications.