Delayed Expansion AGT: Kinodynamic Planning with Application to Tractor-Trailer Parking

2025-06-16

AI Summary

Motion planning for articulated vehicles (e.g., tractor-triple-trailer) in confined, cluttered environments faces challenges including high-dimensional state spaces, nonholonomic constraints, computational inefficiency, and poor goal-reaching accuracy. Method: This paper proposes a learning-augmented, model-based real-time planning framework. It introduces a delayed-expansion motion primitive mechanism to enable online mode classification and priority scheduling; integrates a lightweight learned cost-to-go neural network and a closed-loop trajectory tracking controller to facilitate efficient heuristic search and a novel termination criterion; and unifies precomputed motion primitives, A* tree search, and real-time mode switching. Contribution/Results: Evaluated on autonomous parking tasks, the method achieves a 10× speedup in planning latency, significantly improves success rate and pose accuracy, and— for the first time—enables real-time, precise, and robust parking of multi-segment articulated vehicles in highly constrained spaces.

Technology Category

• Planning, Routing, and Scheduling: Learning for Planning and Scheduling
• Intelligent Robots: Motion and Path Planning
• Multiagent Systems: Multiagent Planning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Kinodynamic planning of articulated vehicles in cluttered environments faces additional challenges arising from high-dimensional state space and complex system dynamics. Built upon [1],[2], this work proposes the DE-AGT algorithm that grows a tree using pre-computed motion primitives (MPs) and A* heuristics. The first feature of DE-AGT is a delayed expansion of MPs. In particular, the MPs are divided into different modes, which are ranked online. With the MP classification and prioritization, DE-AGT expands the most promising mode of MPs first, which eliminates unnecessary computation and finds solutions faster. To obtain the cost-to-go heuristic for nonholonomic articulated vehicles, we rely on supervised learning and train neural networks for fast and accurate cost-to-go prediction. The learned heuristic is used for online mode ranking and node selection. Another feature of DE-AGT is the improved goal-reaching. Exactly reaching a goal state usually requires a constant connection checking with the goal by solving steering problems -- non-trivial and time-consuming for articulated vehicles. The proposed termination scheme overcomes this challenge by tightly integrating a light-weight trajectory tracking controller with the search process. DE-AGT is implemented for autonomous parking of a general car-like tractor with 3-trailer. Simulation results show an average of 10x acceleration compared to a previous method.

Problem

• Kinodynamic planning for articulated vehicles in cluttered environments
• High-dimensional state space and complex system dynamics challenges
• Autonomous parking for tractor-trailer systems with efficiency

Innovation

• Delayed expansion of motion primitives for efficiency
• Supervised learning for cost-to-go heuristic prediction
• Light-weight trajectory tracking for goal-reaching