Planning-oriented Autonomous Driving

By Yi Hu, Jiazhi Yang, Li Chen, Keyu Li, Chonghao Sima, Xizhou Zhu, Siqi Chai, Senyao Du, Tianwei Lin, Wen Wang and 6 more. Computer Vision and Pattern Recognition, 2022

1,470 citations recorded by the source corpus.

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.

Subfields: Artificial Intelligence, Automotive Engineering

View Paper: https://www.semanticscholar.org/paper/fdd7d5b0f6b8641c356e170fd264cd11f70ba657

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Automotive Engineering

Citations in the corpus

DriveLM: Driving with Graph Visual Question Answering (2023, methodology): “Note that the calculation of ADE , FDE and Collision Rate follows the setting used in UniAD [26] but not ST-P3 [25].”
Recent Advancements in End-to-End Autonomous Driving Using Deep Learning: A Survey (2023, methodology): “However, recent methodologies have improved interpretability in End-to-End models by generating auxiliary outputs [7, 13], attention maps [14, 15, 16, 9, 17, 18], and interpretable maps [18, 19, 8, 20, 12, 21].”
End-to-End Autonomous Driving: Challenges and Frontiers (2023, background): “…by decoding the latent feature representations into other meaningful information besides policy prediction, such as semantic segmentation [2], [11], [15], [28], [29], [31], [52], [139], [163], [208], [209], [210], [224], depth estimation [15], [28], [31], [208], [209], object detection…”
GPT-Driver: Learning to Drive with GPT (2023, methodology): “For a fair comparison with other methods, we adopt the evaluation metrics in UniAD Hu et al. (2023) to evaluate our planned trajectories.”
NAVSIM: Data-Driven Non-Reactive Autonomous Vehicle Simulation and Benchmarking (2024, neutral): “(5) UniAD [21] incorporates a wide range of tasks, such as mapping, tracking, motion, and occupancy prediction in a semi-sequential architecture, which processes feature representations through several transformer decoders culminating in a trajectory planning module.”
GAIA-1: A Generative World Model for Autonomous Driving (2023, background): “To achieve this, the development of a robust model of the world is imperative [1] and huge efforts have been made in the past to build such predictive world models for autonomous driving [2, 3, 4, 5, 6].”
LMDrive: Closed-Loop End-to-End Driving with Large Language Models (2023, methodology): “UniAD [18] devised a framework that incorporates full-stack driving tasks and utilizes query-unified interfaces to communicate be-tween different tasks.”
VADv2: End-to-End Vectorized Autonomous Driving via Probabilistic Planning (2024, background): “Existing learning-based planning methods [23, 19, 21, 40, 16, 54] follow a deterministic paradigm to directly regress the action.”
DriveVLM: The Convergence of Autonomous Driving and Large Vision-Language Models (2024, background): “Under our evaluation metric, the additional information is frequently classified as hallucination, consequently leading to lower scores. nuScenes.”
DriveGPT4: Interpretable End-to-End Autonomous Driving Via Large Language Model (2023, background): “To better integrate and optimize the entire system, some approaches pro-pose training the network in an end-to-end manner, eliminating the need for discontinuous intermediate steps [8], [9], [10].”
Driving Into the Future: Multiview Visual Forecasting and Planning with World Model for Autonomous Driving (2023, methodology): “We follow open-loop evaluation metrics [29, 30] for end-to-end planning, including L2 distance from the GT trajectory and the object collision rate.”
Parting with Misconceptions about Learning-based Vehicle Motion Planning (2023, methodology): “As a result, most current work on learned planning only compares to other learned methods, ignoring rule-based baselines [3, 5, 18].”
Vista: A Generalizable Driving World Model with High Fidelity and Versatile Controllability (2024, neutral): “Moreover, most of them are restricted to a specific control modality [54, 61, 90, 125], which hinders the accommodation to prevailing planning algorithms [12, 14, 21, 56, 58, 64] and extension to more applications like decision-making [127] or user interaction [74].”
A Survey on 3D Gaussian Splatting (2024, neutral): “For comprehensiveness, we provide a collection of representative datasets in Table 1 Following the taxonomy outlined Fig.”
DiffusionDrive: Truncated Diffusion Model for End-to-End Autonomous Driving (2024, neutral): “This data-driven ap-⋄ (a) Single mode regression [8, 16, 20].”