Been There, Scanned That: Nostalgia-Driven LiDAR Compression for Self-Driving Cars

TL;DR

DejaView is a novel LiDAR data compression method for autonomous vehicles that leverages long-term redundancies over days and months, achieving high compression ratios with minimal loss.

Contribution

It introduces a long-term redundancy-based compression technique for LiDAR data, exploiting the limited operating area of autonomous vehicles for improved efficiency.

Findings

Achieves a compression factor of 210 with 15 cm error

Utilizes long-term data redundancies over days and months

Demonstrates effectiveness on two months of LiDAR data

Abstract

An autonomous vehicle can generate several terabytes of sensor data per day. A significant portion of this data consists of 3D point clouds produced by depth sensors such as LiDARs. This data must be transferred to cloud storage, where it is utilized for training machine learning models or conducting analyses, such as forensic investigations in the event of an accident. To reduce network and storage costs, this paper introduces DejaView. Although prior work uses interframe redundancies to compress data, DejaView searches for and uses redundancies on larger temporal scales (days and months) for more effective compression. We designed DejaView with the insight that the operating area of autonomous vehicles is limited and that vehicles mostly traverse the same routes daily. Consequently, the 3D data they collect daily is likely similar to the data they have captured in the past. To capture this, the core of DejaView is a diff operation that compactly represents point clouds as delta w.r.t. 3D data from the past. Using two months of LiDAR data, an end-to-end implementation of DejaView can compress point clouds by a factor of 210 at a reconstruction error of only 15 cm.