Optimizing the Placement of Roadside LiDARs for Autonomous Driving

TL;DR

This paper introduces a method to optimize roadside LiDAR placement for autonomous driving by using a greedy algorithm and a perception predictor, supported by a new dataset from CARLA.

Contribution

It presents a novel greedy algorithm for LiDAR placement optimization and a perception predictor trained on a new CARLA-based dataset.

Findings

The proposed method improves perception performance with optimized LiDAR placement.

The perception predictor effectively evaluates placement quality using single-frame point clouds.

The Roadside-Opt dataset facilitates future research in LiDAR placement optimization.

Abstract

Multi-agent cooperative perception is an increasingly popular topic in the field of autonomous driving, where roadside LiDARs play an essential role. However, how to optimize the placement of roadside LiDARs is a crucial but often overlooked problem. This paper proposes an approach to optimize the placement of roadside LiDARs by selecting optimized positions within the scene for better perception performance. To efficiently obtain the best combination of locations, a greedy algorithm based on perceptual gain is proposed, which selects the location that can maximize the perceptual gain sequentially. We define perceptual gain as the increased perceptual capability when a new LiDAR is placed. To obtain the perception capability, we propose a perception predictor that learns to evaluate LiDAR placement using only a single point cloud frame. A dataset named Roadside-Opt is created using the CARLA simulator to facilitate research on the roadside LiDAR placement problem.