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

TL;DR

EI-Drive addresses the gap of realistic network modeling in autonomous-driving simulation by integrating latency and error-prone communications into a CARLA-based platform. It enables multi-agent data fusion through an edge-AI module and a modular pipeline (sensing, perception, planning, control) to study cooperative perception under dynamic traffic and network conditions. The work introduces built-in world scripts and ScenarioRunner/OpenSCENARIO-compatible EIScenarios, and demonstrates via experiments that transmission latency and frame loss can degrade cooperative perception, while RSU- and spectator-assisted data fusion can improve safety; a key metric is $ \text{SR} = \frac{N_{cf}}{N_{total}} \times 100\% $. The platform is open-source and extensible, offering a practical tool for developing cooperative-driving algorithms robust to realistic networking conditions, with potential broad impact on AV research and deployment.

Abstract

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 transmission 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 transmission 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. All code and documents are accessible on our GitHub page: \url{https://ucd-dare.github.io/eidrive.github.io/}.

Figures (6)

• Figure 1: The framework of EI-Drive, which consists of four main components: simulation environment, edge-AI module, modular pipeline, and agent.
• Figure 2: Multi-modal sensors in pipeline scenarios.
• Figure 3: Various object detection methods in pipeline scenarios.
• Figure 4: Cooperative perception in collision avoidance tasks by oracle method.
• Figure 5: Cooperative perception in object detection tasks by oracle method.