Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Assigning Multi-Robot Tasks to Multitasking Robots

Winston Smith, Yu Zhang

TL;DR

The key contribution is a novel task allocation framework that incorporates the consideration of physical constraints introduced by multitasking, in contrast to the existing work where such constraints are largely ignored.

Abstract

One simplifying assumption in existing and well-performing task allocation methods is that the robots are single-tasking: each robot operates on a single task at any given time. While this assumption is harmless to make in some situations, it can be inefficient or even infeasible in others. In this paper, we consider assigning multi-robot tasks to multitasking robots. The key contribution is a novel task allocation framework that incorporates the consideration of physical constraints introduced by multitasking. This is in contrast to the existing work where such constraints are largely ignored. After formulating the problem, we propose a compilation to weighted MAX-SAT, which allows us to leverage existing solvers for a solution. A more efficient greedy heuristic is then introduced. For evaluation, we first compare our methods with a modern baseline that is efficient for single-tasking robots to validate the benefits of multitasking in synthetic domains. Then, using a site-clearing scenario in simulation, we further illustrate the complex task interaction considered by the multitasking robots in our approach to demonstrate its performance. Finally, we demonstrate a physical experiment to show how multitasking enabled by our approach can benefit task efficiency in a realistic setting.

Assigning Multi-Robot Tasks to Multitasking Robots

TL;DR

The key contribution is a novel task allocation framework that incorporates the consideration of physical constraints introduced by multitasking, in contrast to the existing work where such constraints are largely ignored.

Abstract

One simplifying assumption in existing and well-performing task allocation methods is that the robots are single-tasking: each robot operates on a single task at any given time. While this assumption is harmless to make in some situations, it can be inefficient or even infeasible in others. In this paper, we consider assigning multi-robot tasks to multitasking robots. The key contribution is a novel task allocation framework that incorporates the consideration of physical constraints introduced by multitasking. This is in contrast to the existing work where such constraints are largely ignored. After formulating the problem, we propose a compilation to weighted MAX-SAT, which allows us to leverage existing solvers for a solution. A more efficient greedy heuristic is then introduced. For evaluation, we first compare our methods with a modern baseline that is efficient for single-tasking robots to validate the benefits of multitasking in synthetic domains. Then, using a site-clearing scenario in simulation, we further illustrate the complex task interaction considered by the multitasking robots in our approach to demonstrate its performance. Finally, we demonstrate a physical experiment to show how multitasking enabled by our approach can benefit task efficiency in a realistic setting.

Paper Structure

This paper contains 13 sections, 3 theorems, 5 figures.

Table of Contents

  1. Introduction
  2. Related Work
  3. Approach
  4. Running Example
  5. Preliminaries & Prior Work
  6. Problem Definition
  7. Solution Method
  8. Greedy Heuristic
  9. Results
  10. Synthetic Evaluation
  11. Site Clearing
  12. Order Delivery
  13. Conclusions

Key Result

Theorem III.1

$opt(\beta,X)=opt(\beta',X)$.

Figures (5)

  • Figure 1: Running example in ROS with the boxes stacked.
  • Figure 2: Solution ratio as # of tasks (left), robots (center), and CIRs (right) increases in first (top) and second (bottom) setting.
  • Figure 3: Time taken as # of tasks (left), robots (center), and CIRs (right) increases in the first (top) and second (bottom) setting. The solution time for STAMR is not plotted since it required substantially more time than the others.
  • Figure 4: Four site clearing scenarios with one of their optimal solutions illustrated, respectively.
  • Figure 5: The product delivery scenario in its initial state. The entire simulation environment (left). The baseline, midway through execution; note the clumps of robots (center). Our congestion-aware approach (right).

Theorems & Definitions (9)

  • Definition III.1: Constraint Implication Rule (CIR)
  • Definition III.2: Minimally Implying Subset smith
  • Definition III.3: Compatibility smith
  • Definition III.4: TAMPiC
  • Definition III.5
  • Theorem III.1
  • proof
  • Theorem III.2
  • Corollary III.3