Tsukada Lab. Computer Network Group

Enhancing Reliability in Infrastructure-based Collective Perception: A Dual-Channel Hybrid Delivery Approach with Real-Time Monitoring

Yu Asabe, Ehsan Javanmardi, Jin Nakazato, Manabu Tsukada, Hiroshi Esaki

IEEE Open Journal of Vehicular Technology, vol. 5, pp. 1124–1138, 2024年8月30日

3行要約

  • VTC2023-Spring 版 AutowareV2X を拡張ジャーナル化。Dual-Channel Hybrid Delivery を「ReferenceTime 差分による freshness 判定」として体系化。
  • インフラ側 RSU からの CPM 配信について、リアルタイム PDR モニタリングで Wi-Fi/LTE 切替の挙動を可視化。Wi-Fi の RSSI/PDR ヒートマップと LTE Acceptance Rate (LAR) ヒートマップで「片チャネル劣化時の LTE 補完」を定量実証。
  • 結果: Dual-Channel は Wi-Fi 単体比で PDR を大幅改善しつつ E2E latency 約 30 ms を維持。

論文の図

図 2: システム設計とアーキテクチャ:Autoware(左)と、ETSI ITS の通信スタックを持つ AutowareV2X モジュール(右)の接続。
図 2. システム設計とアーキテクチャ:Autoware(左)と、ETSI ITS の通信スタックを持つ AutowareV2X モジュール(右)の接続。出典: Asabe et al., IEEE OJVT 2024(© 2024 The Authors、CC BY-NC-ND 4.0、著者原稿の図版) の図 2AutowareV2X は Autoware の横で独立した ROS 2 ノードとして動き、CPM などのメッセージを無線でやり取りします。自動運転ソフトを改造せずに、共有された認識情報を使えることが分かります。
図 4: デュアルチャネル・ハイブリッド配信:CPM を路車間の直接通信(Channel 1)と携帯電話網(Channel 2)の両方で同時に送る。
図 4. デュアルチャネル・ハイブリッド配信:CPM を路車間の直接通信(Channel 1)と携帯電話網(Channel 2)の両方で同時に送る。出典: Asabe et al., IEEE OJVT 2024(© 2024 The Authors、CC BY-NC-ND 4.0、著者原稿の図版) の図 4直接通信が遮られたり損失が多いときでも、同じ認識メッセージが携帯電話網経由で車に届きます。論文が提案する冗長化の仕組みが分かります。
図 16: フィールド試験での、デュアルチャネル CPM と Wi-Fi のみの CPM のパケット到達率の時間変化。
図 16. フィールド試験での、デュアルチャネル CPM と Wi-Fi のみの CPM のパケット到達率の時間変化。出典: Asabe et al., IEEE OJVT 2024(© 2024 The Authors、CC BY-NC-ND 4.0、著者原稿の図版) の図 16区間 A・B で Wi-Fi のみの到達率は 0.6 を下回りますが、デュアルチャネルは約 0.7 以上を保ちます。Wi-Fi が劣化する場所を携帯電話網が補っていることが分かります。

アブストラクト(原文)

Standalone autonomous vehicles primarily rely on their onboard sensors and may have blind spots or limited situational awareness in complex or dynamic traffic scenarios, leading to difficulties in making safe decisions. Collective perception enables connected autonomous vehicles (CAVs) to overcome the limitations of standalone autonomous vehicles by sharing sensory information with nearby road users. However, unfavorable conditions of the wireless communication medium it uses can lead to limited reliability and reduced quality of service. In this paper, we propose methods for increasing the reliability of collective perception through real-time packet delivery rate monitoring and a dual-channel hybrid delivery approach. We have implemented AutowareV2X, a vehicle-to-everything (V2X) communication module integrated into the autonomous driving (AD) software Autoware. AutowareV2X provides connectivity to the AD stack, enabling end-to-end (E2E) experimentation and evaluation of CAVs. The Collective Perception Service (CPS) was also implemented, allowing the transmission of Collective Perception Messages (CPMs). Our proposed methods using AutowareV2X were evaluated using actual hardware and vehicles in reallife field tests. Results have indicated that the E2E network latency of the perception information sent is around 30 ms, and the AD software can use shared object data to conduct collision avoidance maneuvers. The dual-channel delivery of CPMs enabled the CAV to dynamically select the best CPM from CPMs received from different links, depending on the freshness of their information. This enabled the reliable transmission of CPMs even when there was significant packet loss on one of the transmitting channels.

動画

関連する動画: AutowareV2X: Reliable V2X Communication and Collective Perception for Autonomous Driving

関連する動画: Autowareの協調型自動運転への拡張:認識・計画・連携の一体化を目指して

コード

関連する研究・ソフトウェア

引用方法

Y. Asabe, E. Javanmardi, J. Nakazato, M. Tsukada, and H. Esaki, “Enhancing Reliability in Infrastructure-based Collective Perception: A Dual-Channel Hybrid Delivery Approach with Real-Time Monitoring,” IEEE Open Journal of Vehicular Technology, vol. 5, pp. 1124–1138, 2024, doi: 10.1109/OJVT.2024.3443877.

BibTeX

@article{Asabe2024,
  title         = {Enhancing Reliability in Infrastructure-based Collective Perception: A Dual-Channel Hybrid Delivery Approach with Real-Time Monitoring},
  author        = {Yu Asabe and Ehsan Javanmardi and Jin Nakazato and Manabu Tsukada and Hiroshi Esaki},
  doi           = {10.1109/OJVT.2024.3443877},
  issn          = {2644-1330},
  year          = {2024},
  date          = {2024-08-30},
  urldate       = {2024-08-30},
  journal       = {IEEE Open Journal of Vehicular Technology},
  volume        = {5},
  pages         = {1124--1138},
  abstract      = {Standalone autonomous vehicles primarily rely on their onboard sensors and may have blind spots or limited situational awareness in complex or dynamic traffic scenarios, leading to difficulties in making safe decisions. Collective perception enables connected autonomous vehicles (CAVs) to overcome the limitations of standalone autonomous vehicles by sharing sensory information with nearby road users. However, unfavorable conditions of the wireless communication medium it uses can lead to limited reliability and reduced quality of service. In this paper, we propose methods for increasing the reliability of collective perception through real-time packet delivery rate monitoring and a dual-channel hybrid delivery approach. We have implemented AutowareV2X, a vehicle-to-everything (V2X) communication module integrated into the autonomous driving (AD) software Autoware. AutowareV2X provides connectivity to the AD stack, enabling end-to-end (E2E) experimentation and evaluation of CAVs. The Collective Perception Service (CPS) was also implemented, allowing the transmission of Collective Perception Messages (CPMs). Our proposed methods using AutowareV2X were evaluated using actual hardware and vehicles in reallife field tests. Results have indicated that the E2E network latency of the perception information sent is around 30 ms, and the AD software can use shared object data to conduct collision avoidance maneuvers. The dual-channel delivery of CPMs enabled the CAV to dynamically select the best CPM from CPMs received from different links, depending on the freshness of their information. This enabled the reliable transmission of CPMs even when there was significant packet loss on one of the transmitting channels.},
  keywords      = {ITS},
  pubstate      = {published},
  tppubtype     = {article}
}