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

TL;DR

Lab2Car is proposed, 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, paving the way for quickly deploying and evaluating candidate motion planners in realistic settings.

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.

Figures (8)

• Figure 1: Lab2Car is a plug-and-play wrapper that transforms a rough trajectory sketch into spatiotemporal constraints (a maneuver). The maneuver defines an optimization problem solved by MPC to obtain a final trajectory that is safe, comfortable, and dynamically feasible. This enables rapid deployment and real-world evaluation of experimental planners that lack such properties.
• Figure 2: Anatomy of a maneuver (left) and spline-space illustration (right).
• Figure 3: Lab2Car in real-world scenario. Gray arrowheads denote initial trajectory sketch from experimental planner. Spatiotemporal constraints denoted as in Fig. \ref{['fig:maneuver']}. Connected dark green circles denote predictions for other agents. Red curve denotes final 8-s open-loop trajectory from MPC.
• Figure 4: Lab2Car rescuing bad trajectory sketches in synthetic scenarios. Color coding as in Fig. \ref{['fig:full1']}
• Figure 5: Lab2Car configurations (ablations).