Creating Self-Fixing Autonomous Aerial Vehicles

Autonomous vehicles (AVs) are showing promise in a number of areas, such as package delivery, human transport, forest-fire reconnaissance, and infrastructure inspection. In order to be economically viable, however, human intervention must be minimized. To achieve this, AVs must be able to self-correct mechanical failures; these failures are challenging because the amount of time the machine has to correct them is very small before it will crash.

Assistant Professor Ryan Caverly (Aerospace Engineering) and Assistant Professor Andrew Lamperski (Electrical and Computer Engineering) are working on a project called “Control of Autonomous Aerial Vehicles: A Data-Driven Approach to Bridge the Gap Between Theory and Practice,” that seeks to derive and validate foundational theory for controlling autonomous aerial vehicles with quick, on-the-fly adaptation to failures. The goal is to build a rigorous framework for low-data identification, fault detection, and controller design based on dissipativity theory, with the goal of control recovery within five seconds.

This project recently received a UMII Seed Grant. UMII Seed Grant funds are intended to promote, catalyze, accelerate and advance UMN-based informatics research in areas related to the MnDRIVE initiative, so that UMN faculty and staff are well prepared to compete for longer term external funding opportunities. This Seed Grant falls under the Robotics, Sensors, and Advanced Manufacturing research area of the MnDRIVE initiative.

Research Computing partners:

• University of Minnesota Informatics Institute