EI-Drive:

The growing interest in autonomous driving calls for realistic simulation platforms capable of accurately simulating cooperative perception process in realistic traffic scenarios. Existing studies for cooperative perception often have not accounted for communication latency and errors in real-world environments. To address this gap, we introduce EI-Drive, an edge-AI based autonomous driving simulation platform that integrates advanced cooperative perception with more realistic communication models. Built on the CARLA framework, EI-Drive features new modules for cooperative perception while taking into account communication latency and errors, providing a more realistic platform for evaluating cooperative perception algorithms. In particular, the platform enables vehicles to fuse data from multiple sources, improving situational awareness and safety in complex environments. With its modular design, EI-Drive allows for detailed exploration of sensing, perception, planning, and control in various cooperative driving scenarios. Experiments using EI-Drive demonstrate significant improvements in vehicle safety and performance, particularly in scenarios with complex traffic flow and network conditions.

Integrating cooperative perception with realistic communication models, EI-Drive allows the exploration of how communication latency and errors affect not only cooperative perception but also the overall performance of autonomous vehicles.

EI-Drive has various built-in scenarios tailored for cooperative perception experiments, where the spectator vehicles and RSUs share perception information with the ego vehicle.

EI-Drive has great flexibility to simply apply communication latency and errors to any perception processes, allowing the researches across both communication and autonomous driving. The communication latency and errors not only impair the performance of cooperative perception, but also have negtive influence to the behavior of the ego vehicle.

EI-Drive supports multiple perception methods and multi-modal sensor inputs, greatly enriches the experiment settings.

We conduct extensive experiments related to cooperative perception and communication latency and errors to evaluate their influence on the overall performance.

In collision avoidance, the ego vehicle is affected by cooperative perception with different settings. Here the larger minimal distance indicates better safety. The tables below demonstrate that cooperative perception improves safety, especially the data shared by RSUs, and the communication latency and error impair the performance.

The figures below demonstrate that cooperative perception significantly extends the perception capability of the ego vehicle, especially the data shared by RSUs.

BibTeX

@article{zhou2024ei,
      title={EI-Drive: A Platform for Cooperative Perception with Realistic Communication Models},
      author={Hanchu Zhou, Edward Xie, Wei Shao, Dechen Gao, Michelle Dong, Junshan Zhang},
      journal={arXiv preprint arXiv:2412.09782},
      year={2024},
      url={https://doi.org/10.48550/arXiv.2412.09782}
    }

EI-Drive: A Platform for Cooperative Perception with Realistic Communication Models

Abstract

Seamlessly Simulates Cooperative Perception and Realistic Communication Models

Collision avoidance and traffic detection with cooperative perception

Communication latency and errors

Perception methods and multi-modality

Experiment results

Collision avoidance experiments

Traffic detection experiments

BibTeX