Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

DART: A Compact Platform For Autonomous Driving Research

Lorenzo Lyons, Thijs Niesten, Laura Ferranti

TL;DR

DART introduces a compact, cost-effective 1:10 autonomous-driving testbed designed for single- and multi-vehicle experiments, built on an augmented JetracerPro AI kit. It details a system identification workflow to derive reliable kinematic and dynamic bicycle models, enabling model-based controllers and realistic simulators for hardware validation. The platform is demonstrated through use cases including distributed MPCC, persistent monitoring, vehicle platooning, and curvature-aware CAMPCC, underscoring DART's versatility for multi-robot coordination and advanced motion planning. A publicly available GitHub repository provides building instructions, data, simulation tools, and controllers to support reproducibility and broad access for research groups.

Abstract

This paper presents the design of a research platform for autonomous driving applications, the Delft's Autonomous-driving Robotic Testbed (DART). Our goal was to design a small-scale car-like robot equipped with all the hardware needed for on-board navigation and control while keeping it cost-effective and easy to replicate. To develop DART, we built on an existing off-the-shelf model and augmented its sensor suite to improve its capabilities for control and motion planning tasks. We detail the hardware setup and the system identification challenges to derive the vehicle's models. Furthermore, we present some use cases where we used DART to test different motion planning applications to show the versatility of the platform. Finally, we provide a git repository with all the details to replicate DART, complete with a simulation environment and the data used for system identification.

DART: A Compact Platform For Autonomous Driving Research

TL;DR

DART introduces a compact, cost-effective 1:10 autonomous-driving testbed designed for single- and multi-vehicle experiments, built on an augmented JetracerPro AI kit. It details a system identification workflow to derive reliable kinematic and dynamic bicycle models, enabling model-based controllers and realistic simulators for hardware validation. The platform is demonstrated through use cases including distributed MPCC, persistent monitoring, vehicle platooning, and curvature-aware CAMPCC, underscoring DART's versatility for multi-robot coordination and advanced motion planning. A publicly available GitHub repository provides building instructions, data, simulation tools, and controllers to support reproducibility and broad access for research groups.

Abstract

This paper presents the design of a research platform for autonomous driving applications, the Delft's Autonomous-driving Robotic Testbed (DART). Our goal was to design a small-scale car-like robot equipped with all the hardware needed for on-board navigation and control while keeping it cost-effective and easy to replicate. To develop DART, we built on an existing off-the-shelf model and augmented its sensor suite to improve its capabilities for control and motion planning tasks. We detail the hardware setup and the system identification challenges to derive the vehicle's models. Furthermore, we present some use cases where we used DART to test different motion planning applications to show the versatility of the platform. Finally, we provide a git repository with all the details to replicate DART, complete with a simulation environment and the data used for system identification.
Paper Structure (15 sections, 13 equations, 10 figures, 3 tables)

This paper contains 15 sections, 13 equations, 10 figures, 3 tables.

Table of Contents

  1. Introduction
  2. related works
  3. DART
  4. System identification
  5. Kinematic bicycle model
  6. Longitudinal dynamics
  7. Steering input to steering angle mapping
  8. Actuation delays
  9. Dynamic bicycle model
  10. Use cases
  11. Distributed MPCC distributed_MPC
  12. Persistent monitoring persistent_monitoring
  13. Vehicle platooning
  14. Contouring MPC CAMPCC
  15. Conclusions

Figures (10)

  • Figure 1: DART, a small-scale robotic platform for autonomous driving research.
  • Figure 2: The velocity profile (top) and the estimated resulting force acting on the vehicle (bottom) measured in response to a step throttle input (shaded area).
  • Figure 3: Friction curve fitting results.
  • Figure 4: Motor curve fitting results.
  • Figure 5: Steering input to steering angle static mapping.
  • ...and 5 more figures