Autonomous Robots Lab
  • Home
  • News
  • Research
    • Autonomous Navigation and Exploration
    • Robot Perception
    • Robot Learning
    • Subterranean Robotics
    • Collision-tolerant Aerial Robots
    • Fixed-Wing UAVs
    • Agile and Physical Interaction Control
    • Underwater Autonomy
    • Intelligent Mobility
    • Robotics for Nuclear Sites
    • Autonomous Robots Arena
    • Code
    • Media
    • Research Presentations
    • Projects
  • Publications
  • Group
    • People
    • Research Collaborators
  • Education
    • Introduction to Aerial Robotics >
      • Online Textbook >
        • Modeling >
          • Frame Rotations and Representations
          • Multirotor Dynamics
        • State Estimation >
          • Inertial Sensors
          • Batch Discrete-Time Estimation
          • The Kalman Filter
        • Flight Control >
          • PID Control
          • LQR Control
          • Linear Model Predictive Control
        • Motion Planning >
          • Holonomic Vehicle BVS
          • Dubins Airplane
          • Collision-free Navigation
          • Structural Inspection Path Planning
        • Simulation Tools >
          • Simulations with SimPy
          • MATLAB & Simulink
          • RotorS Simulator >
            • RotorS Simulator Video Examples
      • Lecture Slides
      • Literature and Links
      • RotorS Simulator
      • Student Projects
      • Homework Assignments
      • Independent Study
      • Video Explanations
      • Syllabus
      • Grade Statistics
    • Autonomous Mobile Robot Design >
      • Lecture Slides
      • Semester Projects
      • Code Repository
      • Literature and Links
      • RotorS Simulator
      • Video Explanations
      • Resources for Semester Projects
      • Syllabus
    • Robotics for DDD Applications
    • CS302 - Data Structures
    • Student Projects >
      • Robot Competitions
      • Undergraduate Researchers Needed
      • ConstructionBots - Student Projects
    • EiT TTK4854 - Robotic Ocean Waste Removal
    • Aerial Robotic Autonomy >
      • Breadth Topics
      • Deep-dive Topics
      • Project & Assignments
      • Literature
    • Robotics Seminars
    • Robotics Days
    • Outreach >
      • Drones Demystified! >
        • Lecture Slides
        • Code Repository
        • Video Explanations
        • RotorS Simulator
        • Online Textbook
      • Autonomous Robots Camp >
        • RotorS Simulator
      • Outreach Student Projects
    • BadgerWorks >
      • General Study Links
      • Learn ROS
      • SubT-Edu
  • Resources
    • Autonomous Robots Arena
    • Robot Development Space
  • Contact
Learning-based Exploration Path Planning
 
 
​In this work we present a new methodology on learning-based path planning for autonomous exploration of subterranean environments using aerial robots. Utilizing a recently proposed graph-based path planner as a "training expert" and following an approach relying on the concepts of imitation learning, we derive a trained policy capable of guiding the robot to autonomously explore underground mine drifts and tunnels. The algorithm utilizes only a short window of range data sampled from the onboard LiDAR and achieves an exploratory behavior similar to that of the training expert with a more than an order of magnitude reduction in computational cost, while simultaneously relaxing the need to maintain a consistent and online reconstructed map of the environment. The trained path planning policy is extensively evaluated both in simulation and experimentally within field tests relating to the autonomous exploration of underground mines.

Dataset

Following this link you can download the training/inference dataset for our work on learning-based exploration path planning. 

Video

Proudly powered by Weebly