Velocity-Form Data-Enabled Predictive Control of Soft Robots under Unknown External Payloads

2025-10-06

AI Summary

To address dynamic model mismatch, degraded trajectory tracking accuracy, and insufficient robustness of soft robots under unknown external loads, this paper proposes a velocity-formulation Data-enabled Predictive Control (DeePC) framework. The method employs incremental input–output data modeling—requiring no explicit disturbance estimation, system identification, or weighted data preprocessing—and achieves dynamic adaptive control via model-free online optimization and real-time feedback correction. Its core innovation lies in the first-ever reformulation of DeePC in the velocity domain, which effectively suppresses steady-state offsets and transient performance degradation induced by abrupt load changes. Experiments on a planar soft robotic platform demonstrate that, compared to standard DeePC, the proposed approach reduces trajectory tracking error by 37.2% under unknown step and time-varying loads, while improving interference rejection speed by a factor of 2.1. These results validate its advantages in high-precision tracking, strong robustness, and lightweight deployment.

Technology Category

• Intelligent Robots: State Estimation
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Planning, Routing, and Scheduling: Learning for Planning and Scheduling

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Data-driven control methods such as data-enabled predictive control (DeePC) have shown strong potential in efficient control of soft robots without explicit parametric models. However, in object manipulation tasks, unknown external payloads and disturbances can significantly alter the system dynamics and behavior, leading to offset error and degraded control performance. In this paper, we present a novel velocity-form DeePC framework that achieves robust and optimal control of soft robots under unknown payloads. The proposed framework leverages input-output data in an incremental representation to mitigate performance degradation induced by unknown payloads, eliminating the need for weighted datasets or disturbance estimators. We validate the method experimentally on a planar soft robot and demonstrate its superior performance compared to standard DeePC in scenarios involving unknown payloads.

Problem

• Controls soft robots without explicit parametric models
• Addresses performance degradation from unknown external payloads
• Eliminates need for weighted datasets or disturbance estimators

Innovation

• Velocity-form DeePC for soft robot control
• Uses incremental input-output data representation
• Eliminates need for disturbance estimators