Coupled Sensor Configuration and Planning with Unmanned Aerial Vehicles

Gulden, Jarrett; Nelson, Marina; Letourneau, Jake; Lamar, Thomas; Korfuzi, Peter; Calomo, Joseph; Ballentine, Alexandra

Student Work

Coupled Sensor Configuration and Planning with Unmanned Aerial Vehicles

Pubblico Deposited

The goal of this project is to create an indoor laboratory environment for the collection of multi-modal datasets which may be used to validate path-planning and sensor algorithms. The environment should include various sensors and sensor types, multiple robotic vehicles, and should use a motion capture system for localization. The main outcome of this MQP is a software framework capable of autonomously controlling the robotic vehicles, collecting data, and for wireless and wired communication between relevant entities in the experiment. The environment is relatively easy to operate for novice users, including PhD students in the advisor's lab and is modular to allow for a wide variety of experiments. Data gathered from the MQP should be relevant in validation of coupled sensor configuration and planning algorithms, which means that path planning occurs simultaneously with sensor configuration. Multiple robotic vehicles are developed—specifically wheeled vehicles and unmanned aerial vehicles (UAVs). The wheeled vehicles are solely coded to allow for proper navigation in the lab environment. The UAV is built, tested, and coded by the team. In the final environment, these robotic vehicles, multiple sensor types, and indoor localization are integrated into a software framework which can handle all communication, control, and data collection. A series of unit tests are conducted to validate the environment, code, and software framework. Successful tests demonstrate the feasibility of the software, the accuracy of the implemented control algorithms, and the validity of the UAV design.

This report represents the work of one or more WPI undergraduate students submitted to the faculty as evidence of completion of a degree requirement. WPI routinely publishes these reports on its website without editorial or peer review.

Creator

Subject

Publisher