Safe and Smooth Path Planning for Autonomous Robot Navigation - ON-206

Preferred Disciplines: Computer Science (Masters or PhD)
Company: Clearpath Robotics
Project Length: 4-6 months (1 unit)
Desired start date: May 2019
Location: Kitchener/Waterloo, ON
No. of Positions: 1
Preferences: Intern must be present on site most of the time

About the Company: 

Clearpath provides self-driving vehicle technology and services to over 500 of the world’s most innovative brands. Proprietary hardware, software, and services are delivered through the company’s industrial and research divisions: Clearpath Robotics and OTTO Motors.

We employ a diverse and highly talented team who live and breathe robotics.  We believe that work must have a high “cool” factor and every day should bring new knowledge. We need more passionate people on our team who are willing and able to push the boundaries of robotics into focused and practical applications.

Clearpath is automating the world and we need your help.  Got what it takes?

Project Description:

The GPS waypoint navigation software, developed by Clearpath robotics, enables point to point navigation of robots in outdoor, unstructured, unknown terrains. Since the software does not require prior mapping of the environment, it has to detect and navigate through obstacles as it drives the robot toward the goal point. This requires development of an optimal, fast, and dynamically intelligent motion planner that can modify the reference path of the robot according to the structure of the surrounding area and present obstacles. The goal of this research project is first to evaluate existing obstacle detection and path planning methods. Second, to modify and implement the best of these methods to generate an obstacle free and smooth path for the robot.

Research Objectives:

  • Evaluate the performance of the GPS Waypoint Navigation system developed by Clearpath Robotics
  • Review literature on existing solutions for reliable obstacle detection and avoidance
  • Tune and modify the open source search based path planner (SBPL) to produce an optimal, dynamically feasible, and obstacle free path for Clearpath robots
  • Explore possible options to increase path following speed of the robot via preprocessing the user feeded path in the navigation server  

Methodology:

  • To be discussed with the researcher

Expertise and Skills Needed:

  • Practical knowledge of state estimation, control, and path planning
  • Strong software development skills (C++, Python) and ROS
  • Familiarity with sensors and their error models, particularly LiDAR, GPS, IMU and vision systems
  • Proficiensy with Linux

For more info or to apply to this applied research position, please

  1. Check your eligibility and find more information about open projects
  2. Interested students need to get the approval from their supervisor and send their CV along with a link to their supervisor’s university webpage by applying through the webform.
Program: