Lab2Car: A Versatile Wrapper for Deploying Experimental Planners in Complex Real-world Environments

2024-09-14

arXiv.org

AI Summary

Experimental autonomous driving planners often fail to ensure safety when deployed in complex real-world road environments. Method: This paper proposes a lightweight, nonlinear-optimization-based trajectory encapsulation framework that dynamically refines raw trajectory sketches—generated by arbitrary planners (including both machine learning–based and classical methods)—into executable trajectories satisfying kinematic/dynamic constraints, safety requirements, ride comfort, and dynamic feasibility. Contribution/Results: The framework introduces the first plug-and-play optimization encapsulation paradigm supporting heterogeneous planners, enabling end-to-end real-vehicle deployment of non-certified planners in highly dynamic urban settings (e.g., Las Vegas casino district). Integrated with ROS and adapted to vehicle control interfaces, it demonstrates robust performance in challenging maneuvers—including cut-in, overtaking, and yielding—significantly accelerating on-vehicle planner validation, facilitating rapid iteration, and enabling large-scale evaluation.

Technology Category

• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Search and Optimization: Learning to Search
• Intelligent Robots: Motion and Path Planning

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Agentic search
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Human-level autonomous driving is an ever-elusive goal, with planning and decision making -- the cognitive functions that determine driving behavior -- posing the greatest challenge. Despite a proliferation of promising approaches, progress is stifled by the difficulty of deploying experimental planners in naturalistic settings. In this work, we propose Lab2Car, an optimization-based wrapper that can take a trajectory sketch from an arbitrary motion planner and convert it to a safe, comfortable, dynamically feasible trajectory that the car can follow. This allows motion planners that do not provide such guarantees to be safely tested and optimized in real-world environments. We demonstrate the versatility of Lab2Car by using it to deploy a machine learning (ML) planner and a classical planner on self-driving cars in Las Vegas. The resulting systems handle challenging scenarios, such as cut-ins, overtaking, and yielding, in complex urban environments like casino pick-up/drop-off areas. Our work paves the way for quickly deploying and evaluating candidate motion planners in realistic settings, ensuring rapid iteration and accelerating progress towards human-level autonomy.

Problem

• Deploying experimental planners in real-world driving environments.
• Ensuring safe, comfortable, and feasible trajectories for autonomous vehicles.
• Facilitating rapid testing and optimization of motion planners in complex urban settings.

Innovation

• Optimization-based wrapper for trajectory conversion
• Deploys ML and classical planners in real-world
• Ensures safe, comfortable, dynamically feasible trajectories