A Review of Learning-Based Motion Planning: Toward a Data-Driven Optimal Control Approach

TL;DR

This paper reviews learning-based motion planning for autonomous driving, highlighting the trade-offs and proposing a data-driven optimal control framework to enhance safety, interpretability, and adaptability in autonomous systems.

Contribution

It introduces a unifying data-driven optimal control paradigm that combines classical control structure with machine learning for improved autonomous driving systems.

Findings

Identifies limitations of current learning-based methods.

Proposes a framework integrating control theory with machine learning.

Outlines future directions for safe, interpretable autonomous systems.

Abstract

Motion planning for high-level autonomous driving is constrained by a fundamental trade-off between the transparent, yet brittle, nature of pipeline methods and the adaptive, yet opaque, "black-box" characteristics of modern learning-based systems. This paper critically synthesizes the evolution of the field -- from pipeline methods through imitation learning, reinforcement learning, and generative AI -- to demonstrate how this persistent dilemma has hindered the development of truly trustworthy systems. To resolve this impasse, we conduct a comprehensive review of learning-based motion planning methods. Based on this review, we outline a data-driven optimal control paradigm as a unifying framework that synergistically integrates the verifiable structure of classical control with the adaptive capacity of machine learning, leveraging real-world data to continuously refine key components such as system dynamics, cost functions, and safety constraints. We explore this framework's potential to enable three critical next-generation capabilities: "Human-Centric" customization, "Platform-Adaptive" dynamics adaptation, and "System Self-Optimization" via self-tuning. We conclude by proposing future research directions based on this paradigm, aimed at developing intelligent transportation systems that are simultaneously safe, interpretable, and capable of human-like autonomy.