Figures
The following images are exported plots from the Simscape simulations corresponding to each video shown above. These figures provide detailed analysis of the robot's performance during different locomotion tasks.
In this project, we developed a 2D biped robot model using MATLAB Simscape, focusing on integrating physical constraints and using MPC control to achieve stable locomotion. We began by defining the robot's physical characteristics and configuring the Simscape environment to simulate 2D gravity and contact dynamics, that was crucial for realistic and accurate simulations. Then designed proportional-derivative (PD) controllers to stabilize the robot's joints. we employed linear quadratic regulator (LQR) control strategies to optimize the robot's performance in standing and balancing tasks. The project's core phase was the implementation of Model Predictive Control (MPC), which allowed the robot to anticipate future states and optimize its actions. By integrating MPC with gait scheduling, we enabled the robot to perform dynamic tasks such as walking forward, backward, and stair climbing.
The following videos demonstrate the biped robot's locomotion capabilities through MATLAB Simscape simulations. Each simulation showcases different walking patterns: forward walking, backward walking, and stair climbing. The robot model was designed and simulated using MATLAB Simscape, which provides accurate physics-based modeling for the 2D biped robot with proper contact dynamics and gravity simulation.
The following images are exported plots from the Simscape simulations corresponding to each video shown above. These figures provide detailed analysis of the robot's performance during different locomotion tasks.