OpenCDA: An Open Cooperative Driving Automation Framework Integrated with Co-Simulation

Xu, Runsheng; Guo, Yi; Han, Xu; Xia, Xin; Xiang, Hao; Ma, Jiaqi

doi:10.1109/itsc48978.2021.9564825

Cited by 140 publications

(40 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…OPV2V is an open dataset for perception with V2V communications. It contains 73 scenes collected utilizing CARLA simulator [28] and OpenCDA framework [29]. The dataset is organized as Train, Validate, and Test splits with 6764, 1981, and 2719 frames respectively.…”

Section: Methodsmentioning

confidence: 99%

An Efficient and Robust Object-Level Cooperative Perception Framework for Connected and Automated Driving

Song¹,

Wen²,

Zhang³

et al. 2022

Preprint

View full text Add to dashboard Cite

Cooperative perception is challenging for connected and automated driving because of the real-time requirements and bandwidth limitation, especially when the vehicle location and pose information are inaccurate. We propose an efficient object-level cooperative perception framework, in which data of the 3D bounding boxes, location, and pose are broadcast and received between the connected vehicles, then fused at the object level. Two Iterative Closest Point (ICP) and Optimal Transport theory-based matching algorithms are developed to maximize the total correlations between the 3D bounding boxes jointly detected by the vehicles. Experiment results show that it only takes 5ms to associate objects from different vehicles for each frame, and robust performance is achieved for different levels of location and heading errors. Meanwhile, the proposed framework outperforms the state-of-the-art benchmark methods when location or pose errors occur.

show abstract

Section: Methodsmentioning

confidence: 99%

An Efficient and Robust Object-Level Cooperative Perception Framework for Connected and Automated Driving

Song¹,

Wen²,

Zhang³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…To this end, we collect a new large-scale dataset for V2X perception that explicitly considers these real-world noises during V2X collaboration. We employ CARLA [11] and OpenCDA [42] to generate our dataset. CARLA is responsible for realistic environment rendering and sensor modeling, and OpenCDA offers convenient simultaneous control of multiple AVs and embedded vehicle network communication protocols.…”

Section: V2xset: An Open Dataset For V2x Cooperative Perceptionmentioning

confidence: 99%

V2X-ViT: Vehicle-to-Everything Cooperative Perception with Vision Transformer

Xu¹,

Xiang²,

Tu³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

In this paper, we investigate the application of Vehicle-to-Everything (V2X) communication to improve the perception performance of autonomous vehicles. We present a robust cooperative perception framework with V2X communication using a novel vision Transformer. Specifically, we build a holistic attention model, namely V2X-ViT, to effectively fuse information across on-road agents (i.e., vehicles and infrastructure). V2X-ViT consists of alternating layers of heterogeneous multi-agent self-attention and multi-scale window self-attention, which captures inter-agent interaction and per-agent spatial relationships. These key modules are designed in a unified Transformer architecture to handle common V2X challenges, including asynchronous information sharing, pose errors, and heterogeneity of V2X components. To validate our approach, we create a large-scale V2X perception dataset using CARLA and OpenCDA. Extensive experimental results demonstrate that V2X-ViT sets new state-of-the-art performance for 3D object detection and achieves robust performance even under harsh, noisy environments. The dataset, source code, and trained models will be open-sourced.

show abstract

“…We evaluate the proposed framework on a large-scale open-source multi-agent perception dataset OPV2V [12], which is simulated using the high-fidelity simulator CARLA [43] and a cooperative driving automation simulation framework OpenCDA [44]. It includes 73 scenarios with an average of 25 seconds duration.…”

Section: A Datasetmentioning

confidence: 99%

“…Evaluation metric. Following [44], we evaluate the detection accuracy in the range of x ∈ [−140, 140]m and y ∈ [− 40,40]m, centered at the ego-vehicle coordinate frame. The detection performance is measured with Average Precision (AP) at IoU = 0.7.…”

Section: B Experiments Setupmentioning

confidence: 99%

Model-Agnostic Multi-Agent Perception Framework

Chen¹,

Xu²,

Xiang³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Existing multi-agent perception systems assume that every agent utilizes the same models with identical parameters and architecture, which is often impractical in the real world. The significant performance boost brought by the multi-agent system can be degraded dramatically when the perception models are noticeably different. In this work, we propose a model-agnostic multi-agent framework to reduce the negative effect caused by model discrepancies and maintain confidentiality. Specifically, we consider the perception heterogeneity between agents by integrating a novel uncertainty calibrator which can eliminate the bias among agents' predicted confidence scores. Each agent performs such calibration independently on a standard public database, and therefore the intellectual property can be protected. To further refine the detection accuracy, we also propose a new algorithm called Promotion-Suppression Aggregation (PSA) that considers not only the confidence score of proposals but also the spatial agreement of their neighbors. Our experiments emphasize the necessity of model calibration across different agents, and the results show that our proposed approach outperforms the stateof-the-art baseline methods for 3D object detection on the open OPV2V dataset.

show abstract

OpenCDA: An Open Cooperative Driving Automation Framework Integrated with Co-Simulation

Cited by 140 publications

References 10 publications

An Efficient and Robust Object-Level Cooperative Perception Framework for Connected and Automated Driving

An Efficient and Robust Object-Level Cooperative Perception Framework for Connected and Automated Driving

V2X-ViT: Vehicle-to-Everything Cooperative Perception with Vision Transformer

Model-Agnostic Multi-Agent Perception Framework

Contact Info

Product

Resources

About