Adaptive Model-Base Control of Quadrupeds via Online System Identification using Kalman Filter

2025-06-16

AI Summary

To address the degradation in robustness and tracking accuracy of conventional fixed-parameter model predictive control (MPC) under dynamic load variations, this paper proposes an online adaptive modeling-based MPC framework. The method innovatively employs an extended Kalman filter (EKF) for millisecond-level joint online identification of robot mass and center-of-mass position—significantly improving estimation robustness and real-time performance under strong noise compared to recursive least squares. By integrating dynamics modeling, real-time parameter updating, and MPC co-optimization, the framework enables closed-loop adaptive regulation of model parameters across diverse load conditions. Experimental results demonstrate a 42% reduction in MPC position tracking error and a threefold improvement in convergence stability under high-noise conditions. This work provides a scalable, adaptive solution for high-precision motion control of quadrupedal robots operating under dynamic loading.

Technology Category

• Intelligent Robots: State Estimation
• Machine Learning: Adversarial Learning & Robustness
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Many real-world applications require legged robots to be able to carry variable payloads. Model-based controllers such as model predictive control (MPC) have become the de facto standard in research for controlling these systems. However, most model-based control architectures use fixed plant models, which limits their applicability to different tasks. In this paper, we present a Kalman filter (KF) formulation for online identification of the mass and center of mass (COM) of a four-legged robot. We evaluate our method on a quadrupedal robot carrying various payloads and find that it is more robust to strong measurement noise than classical recursive least squares (RLS) methods. Moreover, it improves the tracking performance of the model-based controller with varying payloads when the model parameters are adjusted at runtime.

Problem

• Online identification of robot mass and COM using Kalman filter
• Improving robustness to measurement noise in payload variations
• Enhancing tracking performance of model-based controllers dynamically

Innovation

• Online system identification using Kalman Filter
• Adaptive control for variable payloads
• Improved tracking with runtime model adjustment