Optimizing Control-Friendly Trajectories with Self-Supervised Residual Learning

2026-01-06

arXiv.org

AI Summary

This work addresses the challenge of achieving high-precision tracking of aggressive trajectories in complex robotic systems, where modeling inaccuracies often impede performance. The authors propose a self-supervised residual learning and trajectory optimization framework that leverages only trajectory-level data to learn unmodeled dynamics as a residual term beyond nominal dynamics, thereby constructing a hybrid dynamical model. This model enables accurate long-horizon predictions at arbitrary integration step sizes and provides analytical gradients suitable for embedding directly into trajectory optimization. By explicitly minimizing the learned residual during optimization, the approach enhances trajectory trackability. Experiments on agile quadrotor flight demonstrate that the generated aggressive trajectories are tracked with high precision by the controller, validating the method’s effectiveness and practicality.

Technology Category

• Search and Optimization: Learning to Search
• Intelligent Robots: Learning & Optimization for ROB
• Multiagent Systems: Adversarial Agents

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Real-world physics can only be analytically modeled with a certain level of precision for modern intricate robotic systems. As a result, tracking aggressive trajectories accurately could be challenging due to the existence of residual physics during controller synthesis. This paper presents a self-supervised residual learning and trajectory optimization framework to address the aforementioned challenges. At first, unknown dynamic effects on the closed-loop model are learned and treated as residuals of the nominal dynamics, jointly forming a hybrid model. We show that learning with analytic gradients can be achieved using only trajectory-level data while enjoying accurate long-horizon prediction with an arbitrary integration step size. Subsequently, a trajectory optimizer is developed to compute the optimal reference trajectory with the residual physics along it minimized. It ends up with trajectories that are friendly to the following control level. The agile flight of quadrotors illustrates that by utilizing the hybrid dynamics, the proposed optimizer outputs aggressive motions that can be precisely tracked.

Problem

• residual dynamics
• trajectory optimization
• control-friendly trajectories
• model inaccuracies
• aggressive trajectory tracking

Innovation

• self-supervised residual learning
• hybrid dynamics
• trajectory optimization
• control-friendly trajectories
• analytic gradients