Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

CoCar NextGen: a Multi-Purpose Platform for Connected Autonomous Driving Research

Marc Heinrich, Maximilian Zipfl, Marc Uecker, Sven Ochs, Martin Gontscharow, Tobias Fleck, Jens Doll, Philip Schörner, Christian Hubschneider, Marc René Zofka, Alexander Viehl, J. Marius Zöllner

TL;DR

CoCar NextGen addresses the need for a flexible, road-legal research vehicle capable of supporting data collection, closed-loop testing, and V2X experimentation across urban to highway scenarios. It achieves this with a modular, standardized hardware-and-software stack, including a dense multi-modal sensor suite (12 LiDARs, 9 cameras, 3 radars), a centralized compute platform, high-bandwidth networking, and an integrated drive-by-wire system. A CARLA digital twin and a sensor data repository enable reproducible validation and external benchmarking, while the design emphasizes extensibility for future sensors and use cases. Collectively, the platform provides a practical path for rigorous real-world evaluation of perception, planning, and control components toward scalable autonomous driving research.

Abstract

Real world testing is of vital importance to the success of automated driving. While many players in the business design purpose build testing vehicles, we designed and build a modular platform that offers high flexibility for any kind of scenario. CoCar NextGen is equipped with next generation hardware that addresses all future use cases. Its extensive, redundant sensor setup allows to develop cross-domain data driven approaches that manage the transfer to other sensor setups. Together with the possibility of being deployed on public roads, this creates a unique research platform that supports the road to automated driving on SAE Level 5.

CoCar NextGen: a Multi-Purpose Platform for Connected Autonomous Driving Research

TL;DR

CoCar NextGen addresses the need for a flexible, road-legal research vehicle capable of supporting data collection, closed-loop testing, and V2X experimentation across urban to highway scenarios. It achieves this with a modular, standardized hardware-and-software stack, including a dense multi-modal sensor suite (12 LiDARs, 9 cameras, 3 radars), a centralized compute platform, high-bandwidth networking, and an integrated drive-by-wire system. A CARLA digital twin and a sensor data repository enable reproducible validation and external benchmarking, while the design emphasizes extensibility for future sensors and use cases. Collectively, the platform provides a practical path for rigorous real-world evaluation of perception, planning, and control components toward scalable autonomous driving research.

Abstract

Real world testing is of vital importance to the success of automated driving. While many players in the business design purpose build testing vehicles, we designed and build a modular platform that offers high flexibility for any kind of scenario. CoCar NextGen is equipped with next generation hardware that addresses all future use cases. Its extensive, redundant sensor setup allows to develop cross-domain data driven approaches that manage the transfer to other sensor setups. Together with the possibility of being deployed on public roads, this creates a unique research platform that supports the road to automated driving on SAE Level 5.
Paper Structure (16 sections, 9 figures, 1 table)

This paper contains 16 sections, 9 figures, 1 table.

Table of Contents

  1. INTRODUCTION
  2. RELATED WORK
  3. DESIGN PROCESS
  4. Sensor Setup Design
  5. HARDWARE SETUP
  6. LiDAR Sensors
  7. Cameras
  8. Radar Sensors
  9. Wireless Communication
  10. Human-Machine Interface
  11. Computing Platform
  12. Network Configuration
  13. Vehicle Interface
  14. Power-Supply
  15. DIGITAL TWIN AND SAMPLE DATA
  16. ...and 1 more sections

Figures (9)

  • Figure 1: CoCar NextGen was presented first on the IEEE ITSC 2023 in Bilbao
  • Figure 2: Comparison of the ground level field-of-view of the different sensors at close range.
  • Figure 3: Example point clouds of each LiDAR modality. \ref{['fig:os1']} and \ref{['fig:os2']} show the long and medium range 360 LiDARs. The colors denote the reflectivity of the objects. \ref{['fig:aeva']} depicts the 4D LiDAR sensors, where the colors represent the radial velocity of each point relative to the ego vehicle.
  • Figure 4: Sensor setup of CoCar NextGen. 4D-LiDAR (cyan), 360°-LiDAR (blue close-range, yellow far-range), Full-HD cameras (orange), Radar (green)
  • Figure 5: Example images of the nine cameras built onto the rooftop structure.
  • ...and 4 more figures