DriveWorld: 4D Pre-trained Scene Understanding via World Models for Autonomous Driving

Vision-centric autonomous driving has recently raised wide attention due to its lower cost. Pre-training is essential for extracting a universal representation. However, current vision-centric pre-training typically relies on either 2D or 3D pre-text tasks, overlooking the temporal characteristics of autonomous driving as a 4D scene understanding task. In this paper, we address this challenge by introducing a world model-based autonomous driving 4D representation learning framework, dubbed \emph{DriveWorld}, which is capable of pre-training from multi-camera driving videos in a spatio-temporal fashion. Specifically, we propose a Memory State-Space Model for spatio-temporal modelling, which consists of a Dynamic Memory Bank module for learning temporal-aware latent dynamics to predict future changes and a Static Scene Propagation module for learning spatial-aware latent statics to offer comprehensive scene contexts. We additionally introduce a Task Prompt to decouple task-aware features for various downstream tasks. The experiments demonstrate that DriveWorld delivers promising results on various autonomous driving tasks. When pre-trained with the OpenScene dataset, DriveWorld achieves a 7.5% increase in mAP for 3D object detection, a 3.0% increase in IoU for online mapping, a 5.0% increase in AMOTA for multi-object tracking, a 0.1m decrease in minADE for motion forecasting, a 3.0% increase in IoU for occupancy prediction, and a 0.34m reduction in average L2 error for planning.

Towards Objectively Benchmarking Social Intelligence for Language Agents at Action Level

Prominent large language models have exhibited human-level performance in many domains, even enabling the derived agents to simulate human and social interactions. While practical works have substantiated the practicability of grounding language agents in sandbox simulation or embodied simulators, current social intelligence benchmarks either stay at the language level or use subjective metrics. In pursuit of a more realistic and objective evaluation, we introduce the Social Tasks in Sandbox Simulation (STSS) benchmark, which assesses language agents \textbf{objectively} at the \textbf{action level} by scrutinizing the goal achievements within the multi-agent simulation. Additionally, we sample conversation scenarios to build a language-level benchmark to provide an economically prudent preliminary evaluation and align with prevailing benchmarks. To gauge the significance of agent architecture, we implement a target-driven planning (TDP) module as an adjunct to the existing agent. Our evaluative findings highlight that the STSS benchmark is challenging for state-of-the-art language agents. Furthermore, it effectively discriminates between distinct language agents, suggesting its usefulness as a benchmark for evaluating both language models and agent architectures.

AgentAvatar: Disentangling Planning, Driving and Rendering for Photorealistic Avatar Agents

In this study, our goal is to create interactive avatar agents that can autonomously plan and animate nuanced facial movements realistically, from both visual and behavioral perspectives. Given high-level inputs about the environment and agent profile, our framework harnesses LLMs to produce a series of detailed text descriptions of the avatar agents' facial motions. These descriptions are then processed by our task-agnostic driving engine into motion token sequences, which are subsequently converted into continuous motion embeddings that are further consumed by our standalone neural-based renderer to generate the final photorealistic avatar animations. These streamlined processes allow our framework to adapt to a variety of non-verbal avatar interactions, both monadic and dyadic. Our extensive study, which includes experiments on both newly compiled and existing datasets featuring two types of agents -- one capable of monadic interaction with the environment, and the other designed for dyadic conversation -- validates the effectiveness and versatility of our approach. To our knowledge, we advanced a leap step by combining LLMs and neural rendering for generalized non-verbal prediction and photo-realistic rendering of avatar agents.

Fast and Accurate Deep Loop Closing and Relocalization for Reliable LiDAR SLAM

Loop closing and relocalization are crucial techniques to establish reliable and robust long-term SLAM by addressing pose estimation drift and degeneration. This article begins by formulating loop closing and relocalization within a unified framework. Then, we propose a novel multi-head network LCR-Net to tackle both tasks effectively. It exploits novel feature extraction and pose-aware attention mechanism to precisely estimate similarities and 6-DoF poses between pairs of LiDAR scans. In the end, we integrate our LCR-Net into a SLAM system and achieve robust and accurate online LiDAR SLAM in outdoor driving environments. We thoroughly evaluate our LCR-Net through three setups derived from loop closing and relocalization, including candidate retrieval, closed-loop point cloud registration, and continuous relocalization using multiple datasets. The results demonstrate that LCR-Net excels in all three tasks, surpassing the state-of-the-art methods and exhibiting a remarkable generalization ability. Notably, our LCR-Net outperforms baseline methods without using a time-consuming robust pose estimator, rendering it suitable for online SLAM applications. To our best knowledge, the integration of LCR-Net yields the first LiDAR SLAM with the capability of deep loop closing and relocalization. The implementation of our methods will be made open-source.

UniWorld: Autonomous Driving Pre-training via World Models

In this paper, we draw inspiration from Alberto Elfes' pioneering work in 1989, where he introduced the concept of the occupancy grid as World Models for robots. We imbue the robot with a spatial-temporal world model, termed UniWorld, to perceive its surroundings and predict the future behavior of other participants. UniWorld involves initially predicting 4D geometric occupancy as the World Models for foundational stage and subsequently fine-tuning on downstream tasks. UniWorld can estimate missing information concerning the world state and predict plausible future states of the world. Besides, UniWorld's pre-training process is label-free, enabling the utilization of massive amounts of image-LiDAR pairs to build a Foundational Model.The proposed unified pre-training framework demonstrates promising results in key tasks such as motion prediction, multi-camera 3D object detection, and surrounding semantic scene completion. When compared to monocular pre-training methods on the nuScenes dataset, UniWorld shows a significant improvement of about 1.5% in IoU for motion prediction, 2.0% in mAP and 2.0% in NDS for multi-camera 3D object detection, as well as a 3% increase in mIoU for surrounding semantic scene completion. By adopting our unified pre-training method, a 25% reduction in 3D training annotation costs can be achieved, offering significant practical value for the implementation of real-world autonomous driving. Codes are publicly available at https://github.com/chaytonmin/UniWorld.

Controlling Character Motions without Observable Driving Source

How to generate diverse, life-like, and unlimited long head/body sequences without any driving source? We argue that this under-investigated research problem is non-trivial at all, and has unique technical challenges behind it. Without semantic constraints from the driving sources, using the standard autoregressive model to generate infinitely long sequences would easily result in 1) out-of-distribution (OOD) issue due to the accumulated error, 2) insufficient diversity to produce natural and life-like motion sequences and 3) undesired periodic patterns along the time. To tackle the above challenges, we propose a systematic framework that marries the benefits of VQ-VAE and a novel token-level control policy trained with reinforcement learning using carefully designed reward functions. A high-level prior model can be easily injected on top to generate unlimited long and diverse sequences. Although we focus on no driving sources now, our framework can be generalized for controlled synthesis with explicit driving sources. Through comprehensive evaluations, we conclude that our proposed framework can address all the above-mentioned challenges and outperform other strong baselines very significantly.

Contrastive Label Disambiguation for Self-Supervised Terrain Traversability Learning in Off-Road Environments

Discriminating the traversability of terrains is a crucial task for autonomous driving in off-road environments. However, it is challenging due to the diverse, ambiguous, and platform-specific nature of off-road traversability. In this paper, we propose a novel self-supervised terrain traversability learning framework, utilizing a contrastive label disambiguation mechanism. Firstly, weakly labeled training samples with pseudo labels are automatically generated by projecting actual driving experiences onto the terrain models constructed in real time. Subsequently, a prototype-based contrastive representation learning method is designed to learn distinguishable embeddings, facilitating the self-supervised updating of those pseudo labels. As the iterative interaction between representation learning and pseudo label updating, the ambiguities in those pseudo labels are gradually eliminated, enabling the learning of platform-specific and task-specific traversability without any human-provided annotations. Experimental results on the RELLIS-3D dataset and our Gobi Desert driving dataset demonstrate the effectiveness of the proposed method.

Traversability Analysis for Autonomous Driving in Complex Environment: A LiDAR-based Terrain Modeling Approach

For autonomous driving, traversability analysis is one of the most basic and essential tasks. In this paper, we propose a novel LiDAR-based terrain modeling approach, which could output stable, complete and accurate terrain models and traversability analysis results. As terrain is an inherent property of the environment that does not change with different view angles, our approach adopts a multi-frame information fusion strategy for terrain modeling. Specifically, a normal distributions transform mapping approach is adopted to accurately model the terrain by fusing information from consecutive LiDAR frames. Then the spatial-temporal Bayesian generalized kernel inference and bilateral filtering are utilized to promote the stability and completeness of the results while simultaneously retaining the sharp terrain edges. Based on the terrain modeling results, the traversability of each region is obtained by performing geometric connectivity analysis between neighboring terrain regions. Experimental results show that the proposed method could run in real-time and outperforms state-of-the-art approaches.

PACER: A Fully Push-forward-based Distributional Reinforcement Learning Algorithm

In this paper, we propose the first fully push-forward-based Distributional Reinforcement Learning algorithm, called Push-forward-based Actor-Critic EncourageR (PACER). Specifically, PACER establishes a stochastic utility value policy gradient theorem and simultaneously leverages the push-forward operator in the construction of both the actor and the critic. Moreover, based on maximum mean discrepancies (MMD), a novel sample-based encourager is designed to incentivize exploration. Experimental evaluations on various continuous control benchmarks demonstrate the superiority of our algorithm over the state-of-the-art.

Occ-BEV: Multi-Camera Unified Pre-training via 3D Scene Reconstruction

Multi-camera 3D perception has emerged as a prominent research field in autonomous driving, offering a viable and cost-effective alternative to LiDAR-based solutions. However, existing multi-camera algorithms primarily rely on monocular image pre-training, which overlooks the spatial and temporal correlations among different camera views. To address this limitation, we propose the first multi-camera unified pre-training framework called Occ-BEV, which involves initially reconstructing the 3D scene as the foundational stage and subsequently fine-tuning the model on downstream tasks. Specifically, a 3D decoder is designed for leveraging Bird's Eye View (BEV) features from multi-view images to predict the 3D geometric occupancy to enable the model to capture a more comprehensive understanding of the 3D environment. A significant benefit of Occ-BEV is its capability of utilizing a considerable volume of unlabeled image-LiDAR pairs for pre-training purposes. The proposed multi-camera unified pre-training framework demonstrates promising results in key tasks such as multi-camera 3D object detection and surrounding semantic scene completion. When compared to monocular pre-training methods on the nuScenes dataset, Occ-BEV shows a significant improvement of about 2.0% in mAP and 2.0% in NDS for multi-camera 3D object detection, as well as a 3% increase in mIoU for surrounding semantic scene completion. Codes are publicly available at https://github.com/chaytonmin/Occ-BEV.