I was a part of the Pitt Robotics & Automation Society (Pitt RAS) team for Mission 7 of the International Aerial Robotics Competition (IARC) in 2017 and 2018. The project was active from June of 2016 through August of 2018. The team developed multiple fully autonomous air vehicles that were entirely custom-built.
The IARC is an annual competition requiring teams to develop aerial robots that solve problems on the cutting edge of what is currently achievable by any aerial robots, whether owned by industry or governments. Mission 7 involved developing an autonomous drone capable of interacting with randomly moving robots on the ground to direct them towards a goal. Additionally, the drone needed to navigate in an indoor environment without GPS or SLAM, meaning that purely optical methods had to be used.
The team developed complex mechanical, electrical, and software systems in order to achieve its goals. In some aspects, the results surpassed what could be found commercially. For the final technical overview, check out the 2018 Technical Postmortem here.
The following autonomous behaviours were demonstrated on the RAS drone in 2018:
- Autonomous flight
- Localizing and remaining within the arena video
- Avoidance of moving obstacles video
- Blocking and landing on moving targets (Not demonstrated at competition) blocking hitting
In 2017, the team won the award for Best System Design and achieved the most points at the American venue of the IARC (Georgia Tech). The other awards were Most Innovative Design, Best Technical Paper, Best Presentation, and Best T-Shirt.
In 2018, the team again won Best System Design and achieved the most points at the American venue. Pitt also won the award for Best Technical Paper.
As part of the team, I co-lead our motion planning team, which involved leading a team of 5-7 people. One of the things I worked on was the development, improvement, and integration of an ARA* serach based planner that handled dynamic constraints. With this planner, the drone was able to follow dynamically feasible, smooth trajectories around obstalces. Test flights can be found here and here.
The Team
