Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

SmartCooper: Vehicular Collaborative Perception with Adaptive Fusion and Judger Mechanism

Yuang Zhang, Haonan An, Zhengru Fang, Guowen Xu, Yuan Zhou, Xianhao Chen, Yuguang Fang

TL;DR

SmartCooper is introduced, an adaptive collaborative perception framework that incorporates communication optimization and a judger mechanism to facilitate CAV data fusion and evaluates the effectiveness of the proposed algorithm on the OpenCOOD platform.

Abstract

In recent years, autonomous driving has garnered significant attention due to its potential for improving road safety through collaborative perception among connected and autonomous vehicles (CAVs). However, time-varying channel variations in vehicular transmission environments demand dynamic allocation of communication resources. Moreover, in the context of collaborative perception, it is important to recognize that not all CAVs contribute valuable data, and some CAV data even have detrimental effects on collaborative perception. In this paper, we introduce SmartCooper, an adaptive collaborative perception framework that incorporates communication optimization and a judger mechanism to facilitate CAV data fusion. Our approach begins with optimizing the connectivity of vehicles while considering communication constraints. We then train a learnable encoder to dynamically adjust the compression ratio based on the channel state information (CSI). Subsequently, we devise a judger mechanism to filter the detrimental image data reconstructed by adaptive decoders. We evaluate the effectiveness of our proposed algorithm on the OpenCOOD platform. Our results demonstrate a substantial reduction in communication costs by 23.10\% compared to the non-judger scheme. Additionally, we achieve a significant improvement on the average precision of Intersection over Union (AP@IoU) by 7.15\% compared with state-of-the-art schemes.

SmartCooper: Vehicular Collaborative Perception with Adaptive Fusion and Judger Mechanism

TL;DR

SmartCooper is introduced, an adaptive collaborative perception framework that incorporates communication optimization and a judger mechanism to facilitate CAV data fusion and evaluates the effectiveness of the proposed algorithm on the OpenCOOD platform.

Abstract

In recent years, autonomous driving has garnered significant attention due to its potential for improving road safety through collaborative perception among connected and autonomous vehicles (CAVs). However, time-varying channel variations in vehicular transmission environments demand dynamic allocation of communication resources. Moreover, in the context of collaborative perception, it is important to recognize that not all CAVs contribute valuable data, and some CAV data even have detrimental effects on collaborative perception. In this paper, we introduce SmartCooper, an adaptive collaborative perception framework that incorporates communication optimization and a judger mechanism to facilitate CAV data fusion. Our approach begins with optimizing the connectivity of vehicles while considering communication constraints. We then train a learnable encoder to dynamically adjust the compression ratio based on the channel state information (CSI). Subsequently, we devise a judger mechanism to filter the detrimental image data reconstructed by adaptive decoders. We evaluate the effectiveness of our proposed algorithm on the OpenCOOD platform. Our results demonstrate a substantial reduction in communication costs by 23.10\% compared to the non-judger scheme. Additionally, we achieve a significant improvement on the average precision of Intersection over Union (AP@IoU) by 7.15\% compared with state-of-the-art schemes.
Paper Structure (18 sections, 5 equations, 4 figures, 2 tables, 1 algorithm)

This paper contains 18 sections, 5 equations, 4 figures, 2 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Collaborative Perception
  4. Which Vehicle to Communicate
  5. Our Proposed Scheme
  6. Channel-aware Optimization
  7. CAV Communication Scheme
  8. Channel-aware Adjuster
  9. Adaptive Judger Mechanism
  10. Disadvantage Analysis
  11. Scoring & Threshold Scheme
  12. Overall Architecture
  13. Experiments
  14. Dataset
  15. Baselines
  16. ...and 3 more sections

Figures (4)

  • Figure 1: Four possible scoring scenarios where the orange represents the Ego CAV, and the green represents CAV1.
  • Figure 2: Overall architecture of SmartCooper consists of three stages. (1) Network optimization: determining the parameters of CSI to output the optimized compression ratio $\alpha^{*}$. (2) Transmission & Judgement: Each CAV transmits its compressed image data to the Ego CAV and receives a judgment. (3) Fusion: the qualified reconstructed images are combined with the Ego CAV's image to generate a BEV through the fusion net.
  • Figure 3: Comparison between SmartCooper and C-AOL for the Minimum Required Bandwidth under different power conditions.
  • Figure 4: The comparison of SmartCooper with four baseline methods in terms of BEV prediction.