Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

MCTS Based Dispatch of Autonomous Vehicles under Operational Constraints for Continuous Transportation

Milan Tomy, Konstantin M. Seiler, Andrew J. Hill

Abstract

Continuous transportation of material in the mining industry is achieved by the dispatch of autonomous haul-trucks with discrete haulage capacities. Recently, Monte Carlo Tree Search (MCTS) was successfully deployed in tackling challenges of long-run optimality, scalability and adaptability in haul-truck dispatch. Typically, operational constraints imposed on the mine site are satisfied by heuristic controllers or human operators independent of the dispatch planning. This article incorporates operational constraint satisfaction into the dispatch planning by utilising the MCTS based dispatch planner Flow-Achieving Scheduling Tree (FAST). Operational constraint violation and satisfaction are modelled as opportunity costs in the combinatorial optimisation problem of dispatch. Explicit cost formulations are avoided by utilising MCTS generator models to derive opportunity costs. Experimental studies with four types of operational constraints demonstrate the success of utilising opportunity costs for constraint satisfaction, and the effectiveness of integrating constraints into dispatch planning.

MCTS Based Dispatch of Autonomous Vehicles under Operational Constraints for Continuous Transportation

Abstract

Continuous transportation of material in the mining industry is achieved by the dispatch of autonomous haul-trucks with discrete haulage capacities. Recently, Monte Carlo Tree Search (MCTS) was successfully deployed in tackling challenges of long-run optimality, scalability and adaptability in haul-truck dispatch. Typically, operational constraints imposed on the mine site are satisfied by heuristic controllers or human operators independent of the dispatch planning. This article incorporates operational constraint satisfaction into the dispatch planning by utilising the MCTS based dispatch planner Flow-Achieving Scheduling Tree (FAST). Operational constraint violation and satisfaction are modelled as opportunity costs in the combinatorial optimisation problem of dispatch. Explicit cost formulations are avoided by utilising MCTS generator models to derive opportunity costs. Experimental studies with four types of operational constraints demonstrate the success of utilising opportunity costs for constraint satisfaction, and the effectiveness of integrating constraints into dispatch planning.
Paper Structure (29 sections, 6 equations, 4 figures, 1 table)

This paper contains 29 sections, 6 equations, 4 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Work
  3. System Description
  4. Problem Definition
  5. FAST
  6. Operational Constraint Satisfaction
  7. Opportunity cost
  8. Generating opportunity costs
  9. Designing violation delay
  10. Battery constraints
  11. Tyre temperature constraints
  12. Capacity constraints
  13. Ratio constraints
  14. FAST with operational constraints -- $\text{FAST}_{\text{OC}}$
  15. Empirical Analysis
  16. ...and 14 more sections

Figures (4)

  • Figure 1: Violation of battery operational constraint $b_i > 10 \space \forall v_i\in \mathcal{H}$ being caused by a sequence of decisions. Simplified state transitions of a single truck $v_i$ is depicted. Truck $v_i$ can choose among haulage tasks $T_1, T_2$ or charge action. Red nodes indicate constraint violation.
  • Figure 2: Road network $\mathcal{R}$ where $\text{L}_i$ are loading and $\text{UL}_i$ unloading stations. The scale of the graph edges reflects the transit duration on each edge.
  • Figure 3: Comparing $\text{FAST}_{\text{OC}}$ with battery constraint against $\text{FAST}_{\text{HC}}$ for varying $f_{hz}$ and fixed $f_{hf}$. Horizon $H\!=\!7f_{hz}$ hrs, Halftime $H_{0.5}\!=\!7f_{hf}$ hrs
  • Figure 4: Comparing $\text{FAST}_{\text{OC}}$ with tyre temperature constraint against $\text{FAST}_{\text{HC}}$ for varying $f_{hz}$ and fixed $f_{hf}$. $H=4f_{hz}$ hrs, $H_{0.5}=4f_{hf}$ hrs.