Planning-oriented Autonomous Driving

TL;DR

This paper proposes UniAD, a unified framework for autonomous driving that integrates perception, prediction, and planning tasks into one network, improving coordination and performance on the nuScenes benchmark.

Contribution

The paper introduces UniAD, a comprehensive, unified model that optimizes all driving tasks towards planning, reducing errors and enhancing task coordination in autonomous driving.

Findings

Outperforms previous state-of-the-art methods on nuScenes benchmark

Effectively integrates multiple driving tasks into a single network

Demonstrates significant improvements through extensive ablations

Abstract

Modern autonomous driving system is characterized as modular tasks in sequential order, i.e., perception, prediction, and planning. In order to perform a wide diversity of tasks and achieve advanced-level intelligence, contemporary approaches either deploy standalone models for individual tasks, or design a multi-task paradigm with separate heads. However, they might suffer from accumulative errors or deficient task coordination. Instead, we argue that a favorable framework should be devised and optimized in pursuit of the ultimate goal, i.e., planning of the self-driving car. Oriented at this, we revisit the key components within perception and prediction, and prioritize the tasks such that all these tasks contribute to planning. We introduce Unified Autonomous Driving (UniAD), a comprehensive framework up-to-date that incorporates full-stack driving tasks in one network. It is exquisitely devised to leverage advantages of each module, and provide complementary feature abstractions for agent interaction from a global perspective. Tasks are communicated with unified query interfaces to facilitate each other toward planning. We instantiate UniAD on the challenging nuScenes benchmark. With extensive ablations, the effectiveness of using such a philosophy is proven by substantially outperforming previous state-of-the-arts in all aspects. Code and models are public.