Home - olinrobotics/gravl GitHub Wiki
GRAVL: Ground Robotic Autonomous Vehicle Lab
Overview
GRAVL is an undergraduate research team within the Olin Robotics Lab working on autonomizing and maintaining a fleet of autonomous ground vehicle platforms for use in both contractual and student research projects. We aim to give all of our platforms autonomous navigation and obstacle avoidance, as well as more mission-specific capabilities and hardware dependent on current lab projects. Additionally, we aim for code developed for any given platform to be easily portable to any other platform, and as such are standardizing on a given informational flow and robot-brain structure.
For a video introduction to the Olin Robotics Lab, see "An Introduction to the Olin Robotics Lab" (link)
Videos
- Fall 2019
- Spring 2019
- Following Tractor-Lidar Test
- Blooper Reel [images/180500_gravl_demo-video.png](https://www.youtube.com/watch?v=rLdPoTPenYo) [images/171200_gravl_impact-hcp-demo-video.png](https://www.youtube.com/watch?v=e5cIqTJsXIo) [images/170700_gravl_impact-hcp-demo-video.png](https://www.youtube.com/watch?v=K46BHT-7SUI)
Projects
Repositories
The following repositories are used and maintained by GRAVL:
- gravl - Low-level firmware and software for running Kubo, our full-scale tractor
- crawler - Low-level firmware and software for running Koda, our sub-scale tractor
- tractor_sim - Simulated tractor for testing, equivalent to virtual low-level firmware and software
- state_controller - High-level software for switching between tractor states and behaviors using a gamepad
- cut_mission - High-level software for autonomous road grading behavior
- blade_nav - High-level software for converting measurements to hitch commands
The following repositories were created but are not currently maintained by GRAVL:
- tractor_sim_packages - A collection of packages for pointcloud generation
- dozer - Repository containing code for tutorial platform
- indoor_localization - Repository containing code for an "indoor GPS" using april tags mounted to the ceiling with a upwards-facing camera on the platform. Designed for Koda
For more information, continue browsing this wiki, or contact project leaders Connor Novak at [email protected] or Amy Phung at [email protected].