Coupled Control Lyapunov Functions for Interconnected Systems, With Application to Quadrupedal Locomotion
Wen-Loong Ma, Noel Csomay-Shanklin, Shishir Kolathaya, Kaveh Akbari Hamed, Aaron D. Ames
- Year
- 2021
- Citations
- 14
Abstract
This letter addresses the problem of formally guaranteeing the stability of interconnected systems with local controllers with a view toward stabilizing quadrupeds viewed as coupled bipeds. In particular, we present a novel framework that views general rigid-body systems as a collection of lower-dimensional systems that are coupled via reaction forces. Stabilizing the corresponding coupled control system can thus be addressed by stabilizing each subsystem coupled through the passive dynamics. The main results of the letter are stability conditions that guarantee convergence for each control subsystem by formulating coupled control Lyapunov functions (CCLFs) using the notion of input-to-state stability (ISS). This theoretical result is illustrated via a simple cart-pole example, where exponential stability is obtained. Next, building on previous results where an 18-DOF quadrupedal robot is decomposed into two interconnected bipedal systems for efficient periodic gait generation, we design model-free quadratic programs (QPs) using the CCLFs to stabilize the continuous dynamics and thus achieve experimental walking and simulated hopping and running on the Vision 60 quadrupedal robot.
Keywords
Related papers
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991
A new optimizer using particle swarm theory
R.C. Eberhart, James Kennedy
2002