eCAV: An Edge-Assisted Evaluation Platform for Connected Autonomous Vehicles
🤖 AI Summary
Current autonomous driving simulation platforms struggle to support high-fidelity evaluation of large-scale, multi-vehicle cooperative scenarios under resource constraints, particularly lacking modeling capabilities for V2X control loops and edge-coordination mechanisms. To address this, we propose an edge-enhanced simulation evaluation platform featuring a novel modular and scalable edge-assisted simulation architecture. It is the first to integrate Vehicle-to-Edge (V2E) control-plane modeling and corresponding algorithm evaluation functionality. The platform combines a lightweight simulation kernel, distributed edge scheduling, multi-granularity sensor abstraction, and real-time control decoupling. Experimental results demonstrate that, in perception-free mode, it supports 256 concurrent vehicles—eight times the state-of-the-art scale; in perception-aware mode, it enables 64 vehicles with sub-800 ms per-step latency—achieving 4× and 1.5× improvements in scalability and execution speed over OpenCDA, respectively.
Application Category
📝 Abstract
As autonomous vehicles edge closer to widespread adoption, enhancing road safety through collision avoidance and minimization of collateral damage becomes imperative. Vehicle-to-everything (V2X) technologies, which include vehicle-to-vehicle (V2V), vehicle-to-infrastructure (V2I), and vehicle-to-cloud (V2C), are being proposed as mechanisms to achieve this safety improvement. Simulation-based testing is crucial for early-stage evaluation of Connected Autonomous Vehicle (CAV) control systems, offering a safer and more cost-effective alternative to real-world tests. However, simulating large 3D environments with many complex single- and multi-vehicle sensors and controllers is computationally intensive. There is currently no evaluation framework that can effectively evaluate realistic scenarios involving large numbers of autonomous vehicles. We propose eCAV -- an efficient, modular, and scalable evaluation platform to facilitate both functional validation of algorithmic approaches to increasing road safety, as well as performance prediction of algorithms of various V2X technologies, including a futuristic Vehicle-to-Edge control plane and correspondingly designed control algorithms. eCAV can model up to 256 vehicles running individual control algorithms without perception enabled, which is $8 imes$ more vehicles than what is possible with state-of-the-art alternatives. %faster than state-of-the-art alternatives that can simulate $8 imes$ fewer vehicles. With perception enabled, eCAV simulates up to 64 vehicles with a step time under 800ms, which is $4 imes$ more and $1.5 imes$ faster than the state-of-the-art OpenCDA framework.
Problem
Research questions and friction points this paper is trying to address.
Lack of scalable platform for large-scale CAV scenario evaluation
High computational cost of simulating multi-vehicle 3D environments
Need for efficient V2X technology performance prediction framework
Innovation
Methods, ideas, or system contributions that make the work stand out.
Edge-assisted platform for CAV evaluation
Modular scalable V2X performance prediction
High-capacity simulation up to 256 vehicles
🔎 Similar Papers
No similar papers found.
