Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Behavior Trees in Industrial Applications: A Case Study in Underground Explosive Charging

Mattias Hallen, Matteo Iovino, Shiva Sander-Tavallaey, Christian Smith

TL;DR

Combining BTs with FSMs can increase the readability and maintainability of the system, especially in the use case studied in this paper, where the full system cannot run autonomously but human supervision and feedback are needed.

Abstract

In industrial applications Finite State Machines (FSMs) are often used to implement decision making policies for autonomous systems. In recent years, the use of Behavior Trees (BT) as an alternative policy representation has gained considerable attention. The benefits of using BTs over FSMs are modularity and reusability, enabling a system that is easy to extend and modify. However, there exists few published studies on successful implementations of BTs for industrial applications. This paper contributes with the lessons learned from implementing BTs in a complex industrial use case, where a robotic system assembles explosive charges and places them in holes on the rock face. The main result of the paper is that even if it is possible to model the entire system as a BT, combining BTs with FSMs can increase the readability and maintainability of the system. The benefit of such combination is remarked especially in the use case studied in this paper, where the full system cannot run autonomously but human supervision and feedback are needed.

Behavior Trees in Industrial Applications: A Case Study in Underground Explosive Charging

TL;DR

Combining BTs with FSMs can increase the readability and maintainability of the system, especially in the use case studied in this paper, where the full system cannot run autonomously but human supervision and feedback are needed.

Abstract

In industrial applications Finite State Machines (FSMs) are often used to implement decision making policies for autonomous systems. In recent years, the use of Behavior Trees (BT) as an alternative policy representation has gained considerable attention. The benefits of using BTs over FSMs are modularity and reusability, enabling a system that is easy to extend and modify. However, there exists few published studies on successful implementations of BTs for industrial applications. This paper contributes with the lessons learned from implementing BTs in a complex industrial use case, where a robotic system assembles explosive charges and places them in holes on the rock face. The main result of the paper is that even if it is possible to model the entire system as a BT, combining BTs with FSMs can increase the readability and maintainability of the system. The benefit of such combination is remarked especially in the use case studied in this paper, where the full system cannot run autonomously but human supervision and feedback are needed.
Paper Structure (16 sections, 5 figures)

This paper contains 16 sections, 5 figures.

Table of Contents

  1. Introduction
  2. Background and Related Work
  3. Behavior Trees
  4. Finite State Machines
  5. Related Work
  6. Description of the Use Case
  7. Application Example
  8. Robotic System
  9. Challenges
  10. Proposed solution
  11. Finite State Machine
  12. Behavior Tree
  13. Outcome
  14. Lessons Learned
  15. Faced Challenges
  16. ...and 1 more sections

Figures (5)

  • Figure 1: Workflow of priming explosives and pumping emulsion into a single drilled hole. This process repeats for every single drilled hole.
  • Figure 2: Robotic system for autonomous charge placement. The truck houses emulsion equipment and a hydraulic crane. This hydraulic crane extends the reach of the primary robot mounted at the end. On the robot platform the smaller secondary manipulator handles detonator and primer assembly.
  • Figure 3: Scheme of the state machine. Each state has a corresponding Behavior Tree that triggers upon entering the state.
  • Figure 4: Scheme of the Charging BT. For simplicity, leaves or fallback behaviors that are not useful to understand the functioning principle of the BT were omitted. The white boxes with a label represent subtrees achieving a subtask described by the label.
  • Figure 5: Images from field tests during 2023.