Wireless Networking for Collaborative Drones

Wireless Networking for Collaborative Drones

Project Description:

This proposal is for continuing the work initiated last year with a graduate assistantship for drone-to-drone data communication. The objective of the project is to develop wireless networks to have drones exchange data directly without a ground station. The network is to be tested on application scenarios in which drones are required to collaborate (i.e., to share information and act upon it) for a particular goal. The goal being considered is finding a target in minimum time within an area that cannot be covered by one single drone alone. The results achieved so far include: a) Training graduate assistant on Python programming for drones. The expected graduation date of the student is May 2018, and the student is planning to do a thesis on this project. b) Training graduate assistant on regulations of the Federal Aviation Administration for small unmanned aerial vehicles (sUAV). The graduate assistant, Dhruvil Darji, is scheduled to take the test for obtaining an sUAV pilot’s license on April 14, 2017. The license is required for flying drones for non-recreational purposes. c) Consultation with LMU’s Risk Manager, Mr. Douglas J. Moore, on permission to fly drones on the LMU campus for research purposes. Due to the close proximity to LAX, it is recommended that drones be flown within an outdoor netted facility. Otherwise, permission from LAX air traffic control is required prior to flying drones on the LMU campus. The EECS department is considering the purchase of outdoor netting that can be assembled and disassembled as needed. d) Interfacing wireless radios with the drone’s USB port and programming the radio from the Python scripts developed by the graduate assistant e) Controlling the wireless radios’ transmission power and detecting the corresponding received signal power on the receiving radio f) Obtaining drone’s GPS location and transmitting it using the wireless radio g) Automated generation of Matlab scripts for processing of data for the design of wireless networking protocols for drone-to-drone communication. The data includes GPS location of drone, transmission power, received signal power, and corresponding timing. The project uses results from previously Seaver funded projects on transmission-power control on wireless body networks for motion capture. This work was published in the IEEE World Forum on Internet of Things in 2016. Also, the project requests access to the open-source drones that were approved by the Seaver Equipment Committee for the Department of Electrical Engineering and 2/2 Computer Science in 2016. It is expected that the results of the project will be integrated in educational activities in ELEC-402 Senior Project, ELEC-598 SS:Wireless Networks, and ELEC-698 Wireless Sensor Networks. Drones are currently being used for a wide variety of applications in the filming, entertainment, defense, and transportation industries. In many of the applications, drones are individual systems that operate independently from other drones. For example, an operator controls one single drone to film an event or to participate in a drone-race competition. However, as application scenarios become more complex, several drones may be considered that work collaboratively to perform one specific task. The technical problem addressed in this proposal considers this collaborative scenario. Specifically, the problem consists of a target search in minimum time. When more than one drone is considered, the search time can be reduced because different search regions can be covered simultaneously by different drones. Therefore, in order to minimize the search time, the distance between drones increases. From a wireless networking perspective, the increase in distance is a challenge because the transmission power may have to be increased at levels that may not be feasible when the network is to be maintained and fully connected. Infeasibility may be due to different reasons such as required transmission power that exceeds specifications or that depletes drones’ batteries too fast. This project aims at developing transmission-power and information-routing algorithms that enable minimization of search time while maintaining the necessary network connectivity and routing of information so that drones can share information reliably, with low delay, and reduced energy consumption.

Advisor: Gustavo Vejarano, gvejaran@lmu.edu
Project Duration: 1 Year

Student Duties:

The student will be responsible for the design, implementation, and testing of algorithms that allow the automation of:

  • Collaborative target-search with open-source drones
  • Dynamic transmission power control that adapts to drones' relative positions
  • Dynamic routing of information between drones performing collaborative target search

Student Background Required:

Student should possess programming experience in any language and have a background in computer engineering, electrical engineering, or computer science.