The Surprising Effectiveness of Linear Models for Whole-Body Model-Predictive Control
When do locomotion controllers require reasoning about nonlinearities? In this work, we show that a whole-body model-predictive controller using a simple linear time-invariant approximation of the whole-body dynamics is able to execute basic locomotion tasks on complex legged robots. The formulation requires no online nonlinear dynamics evaluations or matrix inversions. We demonstrate walking, disturbance rejection, and even navigation to a goal position without a separate footstep planner on a quadrupedal robot. In addition, we demonstrate dynamic walking on a hydraulic humanoid, a robot with significant limb inertia, complex actuator dynamics, and large sim-to-real gap.
Check out our website.
Related Papers
|
2025 September |
The Surprising Effectiveness of Linear Models for Whole-Body Model-Predictive Control
IEEE Humanoids. Seoul, South Korea. (Accepted) |
People
Arun Bishop
Contact-rich Optimization and Control
Juan Alvarez Padilla
Learning and Optimization Based Controls
Sam Schoedel
Apptronik
Ibrahima S. Sow
Tesla AI
Will Kraus
Roboligent
Zac Manchester
Associate Professor