Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Scalable Supervisory Architecture for Autonomous Race Cars

Zalán Demeter, Péter Bogdán, Ármin Bogár-Németh, Gergely Bári

TL;DR

The paper addresses the need for a scalable, reusable software architecture for autonomous racing that can support multiple driving agents in parallel across diverse platforms. It presents a modular supervisory framework comprising pipelines, handlers, and vehicle-agnostic interfaces, underpinned by N-self-checking programming and a distributed master-slave setup to enable fault-tolerant parallel execution. The approach is validated in both real-world (F1Tenth IROS23) and simulated (Indy Autonomous Challenge Simulation Race 2023) racing environments, demonstrating consistency, scalability, and cross-domain portability from 1/10-scale to simulated full-scale vehicles. This work offers a practical foundation for rapid development, testing, and deployment of competitive autonomous racing systems across leagues, reducing cost and enhancing reusability of pipeline components.

Abstract

In recent years, the number and importance of autonomous racing leagues, and consequently the number of studies on them, has been growing. The seamless integration between different series has gained attention due to the scene's diversity. However, the high cost of full scale racing makes it a more accessible development model, to research at smaller form factors and scale up the achieved results. This paper presents a scalable architecture designed for autonomous racing that emphasizes modularity, adaptability to diverse configurations, and the ability to supervise parallel execution of pipelines that allows the use of different dynamic strategies. The system showcased consistent racing performance across different environments, demonstrated through successful participation in two relevant competitions. The results confirm the architecture's scalability and versatility, providing a robust foundation for the development of competitive autonomous racing systems. The successful application in real-world scenarios validates its practical effectiveness and highlights its potential for future advancements in autonomous racing technology.

Scalable Supervisory Architecture for Autonomous Race Cars

TL;DR

The paper addresses the need for a scalable, reusable software architecture for autonomous racing that can support multiple driving agents in parallel across diverse platforms. It presents a modular supervisory framework comprising pipelines, handlers, and vehicle-agnostic interfaces, underpinned by N-self-checking programming and a distributed master-slave setup to enable fault-tolerant parallel execution. The approach is validated in both real-world (F1Tenth IROS23) and simulated (Indy Autonomous Challenge Simulation Race 2023) racing environments, demonstrating consistency, scalability, and cross-domain portability from 1/10-scale to simulated full-scale vehicles. This work offers a practical foundation for rapid development, testing, and deployment of competitive autonomous racing systems across leagues, reducing cost and enhancing reusability of pipeline components.

Abstract

In recent years, the number and importance of autonomous racing leagues, and consequently the number of studies on them, has been growing. The seamless integration between different series has gained attention due to the scene's diversity. However, the high cost of full scale racing makes it a more accessible development model, to research at smaller form factors and scale up the achieved results. This paper presents a scalable architecture designed for autonomous racing that emphasizes modularity, adaptability to diverse configurations, and the ability to supervise parallel execution of pipelines that allows the use of different dynamic strategies. The system showcased consistent racing performance across different environments, demonstrated through successful participation in two relevant competitions. The results confirm the architecture's scalability and versatility, providing a robust foundation for the development of competitive autonomous racing systems. The successful application in real-world scenarios validates its practical effectiveness and highlights its potential for future advancements in autonomous racing technology.
Paper Structure (26 sections, 9 figures)

This paper contains 26 sections, 9 figures.

Table of Contents

  1. Introduction
  2. Overview of Autonomous Race car architectures
  3. Architecture Overview
  4. Driving Factors
  5. Components of the architecture
  6. Data Flow
  7. Control Flow
  8. Handlers
  9. Driver Handler
  10. Module Handler
  11. Pipeline Handler
  12. Modules
  13. Vehicle State Estimator
  14. Vehicle Adapter
  15. AS State Machine
  16. ...and 11 more sections

Figures (9)

  • Figure 1: Sense-Think-Act Paradigm: Designing software architecture using this paradigm involves environment perception, processing information, and taking actions based on analysis.
  • Figure 2: High Level Overview: example of multiple pipelines working together in the architecture.
  • Figure 3: Architecture Diagram: Data and control flow of the system. The components depicted are detailed in \ref{['sec:modules']}, \ref{['sec:handlers']} and \ref{['sec:pipelines']}.
  • Figure 4: N-self-checking programming: concurrently operating multiple independent versions with checkers for fault tolerance. This is an example configuration, the depicted components are detailed in \ref{['sec:decisionlogic']} and \ref{['sec:pipelinehandler']}.
  • Figure 5: Software Layer Diagram: shows the hierarchy between functional layers in the system.
  • ...and 4 more figures