AV4EV: Open-Source Modular Autonomous Electric Vehicle Platform for Making Mobility Research Accessible (2312.00951v2)

Abstract: When academic researchers develop and validate autonomous driving algorithms, there is a challenge in balancing high-performance capabilities with the cost and complexity of the vehicle platform. Much of today's research on autonomous vehicles (AV) is limited to experimentation on expensive commercial vehicles that require large skilled teams to retrofit the vehicles and test them in dedicated facilities. On the other hand, 1/10th-1/16th scaled-down vehicle platforms are more affordable but have limited similitude in performance and drivability. To address this issue, we present the design of a one-third-scale autonomous electric go-kart platform with open-source mechatronics design along with fully functional autonomous driving software. The platform's multi-modal driving system is capable of manual, autonomous, and teleoperation driving modes. It also features a flexible sensing suite for the algorithm deployment across perception, localization, planning, and control. This development serves as a bridge between full-scale vehicles and reduced-scale cars while accelerating cost-effective algorithmic advancements. Our experimental results demonstrate the AV4EV platform's capabilities and ease of use for developing new AV algorithms. All materials are available at AV4EV.org to stimulate collaborative efforts within the AV and electric vehicle (EV) communities.

• The paper introduces a modular mechatronic design integrating subsystems like throttle-by-wire and steer-by-wire to emulate full-sized vehicle dynamics.
• The paper demonstrates advanced sensing and ROS2-based open-source algorithms for precise trajectory tracking and dynamic obstacle avoidance.
• The paper validates its platform through competitive testing and provides comprehensive resources to foster collaborative, accessible autonomous mobility research.

Introduction

Autonomous vehicles (AV) represent a rapidly advancing area of technology with significant research efforts dedicated to improving their safety and performance. Accessibility of appropriate testing platforms for research purposes is a critical aspect. Large-scale testing facilities can be costly and hazardous, often excluding academic institutions from contributing effectively to this field of paper. Scaled-down models offer accessibility, but they frequently lack the dynamics and control capabilities of their full-size counterparts. A newly proposed platform aims to mitigate this challenge.

Modular Mechatronic Design

The development of the autonomous electric go-kart centers around a modular mechatronic system, which is fragmented into distinct subsystems such as Power Distribution, Main Control, Throttle-by-Wire, Steer-by-Wire, and Electronic Braking Systems. Each subsystem is coordinated through a Controller Area Network (CAN) for efficient data exchange, mirroring current vehicle design protocols. The Main Control System (MCS) in particular interfaces user commands with the go-kart's mechatronics, facilitating three modes of operation: manual, teleoperated, and autonomous. This go-kart successfully integrates the practicality of a full-sized vehicle while maintaining the advantages of a reduced-scale model.

Sensing and Autonomous Software

A key feature of the autonomous go-kart is its adaptable sensing assembly, designed to cater to a variety of research objectives and modified as necessary. It boasts a LiDAR and a multi-capability OAK-D camera mounted on the rear shelf, alongside a global positioning system and an inertial measurement unit, all of which feed into an onboard laptop for processing. These elements aid in tasks like perception and localization. Alongside this powerful sensor setup, the platform utilizes open-source software based on the Robot Operating System (ROS2), adopting algorithms for both pre-mapped racing and reactive racing situations. It demonstrates the vehicle's capabilities by employing an adaptive pure pursuit controller for high precision trajectory following and a follow-the-gap algorithm for dynamic obstacle avoidance.

Experimental Validation and Community Impact

This platform's effectiveness has been validated through its championship-winning performance at the Autonomous Karting Series Purdue Grand Prix. The comprehensive resources released, including design blueprints, software repositories, demonstration videos, and a detailed bill of materials, are aimed at propelling collaborative research forward. The platform’s modular nature not only simplifies tweaking and upgrading, but it also promotes research in areas such as human-machine interaction, leaning towards the future exploration of dynamic cooperative control. Acknowledgements recognize contributions and support from various entities, emphasizing collaborative advancement in the autonomous go-kart community.

This electric go-kart platform bridges a significant gap, providing researchers with a practical, accessible, and customizable vehicle to further AV advancements, making it a beacon for educational institutions and fostering innovation in the autonomous driving space.