Perceptual Motor Learning with Active Inference Framework for Robust Lateral Control

2025-03-03

AI Summary

This paper addresses key challenges in lateral control for high-level autonomous vehicles operating in dynamic environments—namely, poor adaptability, weak generalization, and high computational overhead. We propose an end-to-end framework integrating perceptuomotor learning with active inference. For the first time, we embed human motor learning principles into a deep generative model, optimizing for minimal prediction error (i.e., “surprise”) as the sole objective—enabling perception-action tight coupling and adaptive lane-keeping without explicit reward signals. Our method combines variational inference, dynamic belief updating, and differentiable motor representation learning, supporting low-data dependency and zero-shot deployment. Evaluations in CARLA demonstrate: (1) lateral control accuracy on par with conventional methods; (2) significantly enhanced environmental adaptability; (3) over 60% reduction in required training data; and (4) no increase in computational cost.

Technology Category

• Search and Optimization: Learning to Search
• Planning, Routing, and Scheduling: Learning for Planning and Scheduling
• Intelligent Robots: Learning & Optimization for ROB

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

This paper presents a novel Perceptual Motor Learning (PML) framework integrated with Active Inference (AIF) to enhance lateral control in Highly Automated Vehicles (HAVs). PML, inspired by human motor learning, emphasizes the seamless integration of perception and action, enabling efficient decision-making in dynamic environments. Traditional autonomous driving approaches--including modular pipelines, imitation learning, and reinforcement learning--struggle with adaptability, generalization, and computational efficiency. In contrast, PML with AIF leverages a generative model to minimize prediction error ("surprise") and actively shape vehicle control based on learned perceptual-motor representations. Our approach unifies deep learning with active inference principles, allowing HAVs to perform lane-keeping maneuvers with minimal data and without extensive retraining across different environments. Extensive experiments in the CARLA simulator demonstrate that PML with AIF enhances adaptability without increasing computational overhead while achieving performance comparable to conventional methods. These findings highlight the potential of PML-driven active inference as a robust alternative for real-world autonomous driving applications.

Problem

• Enhances lateral control in Highly Automated Vehicles using Active Inference.
• Improves adaptability and generalization in autonomous driving systems.
• Reduces computational overhead and data requirements for lane-keeping.

Innovation

• Integrates Perceptual Motor Learning with Active Inference
• Uses generative model to minimize prediction error
• Unifies deep learning with active inference principles