Exploring deep learning based robot perception techniques for navigating outdoor terrains

Abstract

For a ground robot to autonomously navigate to its objectives, it needs to detect and recognize its surroundings and objects therein. From its sensory input, the robot’s AI has to semantically segment the scenes such as terrain, vegetation, man made structures, debris, water-streams, etc. The onboard perception then has to assess and determine intelligently what parts of the scene can be traversed safely to the objective. Currently there is no ground robot capable of autonomously navigating in natural or unstructured environments without any human intervention. The goal of this project is to develop a novel method of vision based perception for assessing navigability of terrains that may be encountered by an autonomous ground vehicle traversing in natural or manmade environments.

Document Details

Document Type
DoD Grant Award
Publication Date
Oct 20, 2022
Source ID
FA23861914001

Entities

People

  • Hanseok Ko

Organizations

  • Air Force Office of Scientific Research
  • Korea University
  • United States Air Force

Tags

Fields of Study

  • Computer science

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Computer Vision.
  • Robotics and Automation.

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems
  • Autonomy