Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

CoBRA: A Composable Benchmark for Robotics Applications

Matthias Mayer, Jonathan Külz, Matthias Althoff

TL;DR

A benchmark suite encompassing a unified format for robots, environments, and task descriptions is introduced, especially useful for modular robots, where the multitude of robots that can be assembled creates a host of additional parameters to optimize.

Abstract

Selecting an optimal robot, its base pose, and trajectory for a given task is currently mainly done by human expertise or trial and error. To evaluate automatic approaches to this combined optimization problem, we introduce a benchmark suite encompassing a unified format for robots, environments, and task descriptions. Our benchmark suite is especially useful for modular robots, where the multitude of robots that can be assembled creates a host of additional parameters to optimize. We include tasks such as machine tending and welding in synthetic environments and 3D scans of real-world machine shops. All benchmarks are accessible through https://cobra.cps.cit.tum.de, a platform to conveniently share, reference, and compare tasks, robot models, and solutions.

CoBRA: A Composable Benchmark for Robotics Applications

TL;DR

A benchmark suite encompassing a unified format for robots, environments, and task descriptions is introduced, especially useful for modular robots, where the multitude of robots that can be assembled creates a host of additional parameters to optimize.

Abstract

Selecting an optimal robot, its base pose, and trajectory for a given task is currently mainly done by human expertise or trial and error. To evaluate automatic approaches to this combined optimization problem, we introduce a benchmark suite encompassing a unified format for robots, environments, and task descriptions. Our benchmark suite is especially useful for modular robots, where the multitude of robots that can be assembled creates a host of additional parameters to optimize. We include tasks such as machine tending and welding in synthetic environments and 3D scans of real-world machine shops. All benchmarks are accessible through https://cobra.cps.cit.tum.de, a platform to conveniently share, reference, and compare tasks, robot models, and solutions.
Paper Structure (11 sections, 6 equations, 1 figure, 3 tables)

This paper contains 11 sections, 6 equations, 1 figure, 3 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Robot (Task) Description
  4. Typical Robotic Tasks
  5. Optimizing Modular Robots
  6. Contributions
  7. Task Description and Problem Statement
  8. Poses
  9. Hybrid Motion Planning Problem
  10. Constraints
  11. Goals

Figures (1)

  • Figure 1: A robot solving the machine tending task Liu2020/Case2b/Schunk_ IWB_CTX_300_linear_0 minimizing mechanical energy. The robot is assembled with the module set $\mathcal{R} = \mathtt{IMPROV}$ in the order $M = [1, 21, 4, 22, 5, 23, 12]$ and its end-effector follows the cyan path $\mathbf{T}_{\mathrm{eef}}$. An animation is available at https://cobra.cps.cit.tum.de/ICRA24/example.