Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Navigating Intelligence: A Survey of Google OR-Tools and Machine Learning for Global Path Planning in Autonomous Vehicles

Alexandre Benoit, Pedram Asef

TL;DR

This paper frames ROMIE's global path planning for autonomous mining sampling as a TSP and benchmarks Google OR-Tools against reinforcement learning methods (Q-Learning and Double Q-Learning) across datasets ranging from 20 to 1000 points. The main approach combines traditional optimization with learning-based strategies, demonstrating that Q-Learning achieves the smallest average deviation from the optimum (≈1.2%) on the tested datasets and highlighting regime-dependent performance where RL excels on smaller problems and OR-Tools LS methods catch up or surpass RL on larger instances. A practical ROMIE web interface built with Django and Google Maps enables real-time, interactive GPP and demonstrates integration between planning and execution components (GPP-LPP). The study provides the first comprehensive comparison of GOT algorithms and RL in a real-world GPP setting, validated on GPS data from mining sites, and points to future directions including deep learning-based solvers to further improve scalability and robustness.

Abstract

We offer a new in-depth investigation of global path planning (GPP) for unmanned ground vehicles, an autonomous mining sampling robot named ROMIE. GPP is essential for ROMIE's optimal performance, which is translated into solving the traveling salesman problem, a complex graph theory challenge that is crucial for determining the most effective route to cover all sampling locations in a mining field. This problem is central to enhancing ROMIE's operational efficiency and competitiveness against human labor by optimizing cost and time. The primary aim of this research is to advance GPP by developing, evaluating, and improving a cost-efficient software and web application. We delve into an extensive comparison and analysis of Google operations research (OR)-Tools optimization algorithms. Our study is driven by the goal of applying and testing the limits of OR-Tools capabilities by integrating Reinforcement Learning techniques for the first time. This enables us to compare these methods with OR-Tools, assessing their computational effectiveness and real-world application efficiency. Our analysis seeks to provide insights into the effectiveness and practical application of each technique. Our findings indicate that Q-Learning stands out as the optimal strategy, demonstrating superior efficiency by deviating only 1.2% on average from the optimal solutions across our datasets.

Navigating Intelligence: A Survey of Google OR-Tools and Machine Learning for Global Path Planning in Autonomous Vehicles

TL;DR

This paper frames ROMIE's global path planning for autonomous mining sampling as a TSP and benchmarks Google OR-Tools against reinforcement learning methods (Q-Learning and Double Q-Learning) across datasets ranging from 20 to 1000 points. The main approach combines traditional optimization with learning-based strategies, demonstrating that Q-Learning achieves the smallest average deviation from the optimum (≈1.2%) on the tested datasets and highlighting regime-dependent performance where RL excels on smaller problems and OR-Tools LS methods catch up or surpass RL on larger instances. A practical ROMIE web interface built with Django and Google Maps enables real-time, interactive GPP and demonstrates integration between planning and execution components (GPP-LPP). The study provides the first comprehensive comparison of GOT algorithms and RL in a real-world GPP setting, validated on GPS data from mining sites, and points to future directions including deep learning-based solvers to further improve scalability and robustness.

Abstract

We offer a new in-depth investigation of global path planning (GPP) for unmanned ground vehicles, an autonomous mining sampling robot named ROMIE. GPP is essential for ROMIE's optimal performance, which is translated into solving the traveling salesman problem, a complex graph theory challenge that is crucial for determining the most effective route to cover all sampling locations in a mining field. This problem is central to enhancing ROMIE's operational efficiency and competitiveness against human labor by optimizing cost and time. The primary aim of this research is to advance GPP by developing, evaluating, and improving a cost-efficient software and web application. We delve into an extensive comparison and analysis of Google operations research (OR)-Tools optimization algorithms. Our study is driven by the goal of applying and testing the limits of OR-Tools capabilities by integrating Reinforcement Learning techniques for the first time. This enables us to compare these methods with OR-Tools, assessing their computational effectiveness and real-world application efficiency. Our analysis seeks to provide insights into the effectiveness and practical application of each technique. Our findings indicate that Q-Learning stands out as the optimal strategy, demonstrating superior efficiency by deviating only 1.2% on average from the optimal solutions across our datasets.

Paper Structure

This paper contains 18 sections, 7 equations, 18 figures, 7 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Related Works
  3. Optimization Methods for solving the Traveling Salesman Problem
  4. Traveling Salesman Problem
  5. Google OR-Tools
  6. RL: Reinforcement Learning
  7. Results Discussion: The Difference Between Local and Global Optimum
  8. Experimental Results and Validation
  9. Experiment Set-up
  10. Hardware Set-up
  11. Software language
  12. Webframe work
  13. Data set
  14. Google OR-Tools VS RL comparative table
  15. Google OR-Tools Implementation and Results
  16. ...and 3 more sections

Figures (18)

  • Figure 1: Diagram of hierarchical sampling processes before undergoing mining activities.
  • Figure 2: ROMIE’s functioning procedure.
  • Figure 3: Overall structure for ROMIE; LPP: Local Path Planning.
  • Figure 4: Overall interaction from GPP to LPP
  • Figure 9: Comparison between HC and SA by implementing Eq. 7. The axes are identical in all directions and represent a straightforward metric ranging from 1 to -1, indicating the length of the mesh.
  • ...and 13 more figures