Kevin Chang

Participant: PROMISE AGEP Research Symposium


Kevin Chang
: Mechanical Engineering Department
Institution: University of Maryland Baltimore County (UMBC)



Motion Capture and Control of a Nano-Quadcopter

Motion capture control of quadrotors is a relatively well known and established method of researching quadcopter flight dynamics. However, these capture systems are usually very expensive because they require many cameras, a large space, and relatively large quadrotors. In this project, we explore the viability of using a minimum sized motion capture camera setup to serve as a framework for autonomous flight of a quadrotor. The system that we utilize consists of 4 Optitrack Motion capture cameras, a Crazyflie 2.0 Nano-quadrotor, and a Pixhawk flight controller. The Optitrack cameras capture the precise position of the quadrotor with in a predefined capture volume. The positional information is sent to a ground station computer via Ethernet. The positional data is then processed and sent wirelessly to the quadcopter. This system will serve as a proof of concept that smaller camera setups are viable. In addition, the system will be used as the foundation for researching various control algorithms for quadrotors.



Energy consumption modeling and estimation for electric unmanned aerial systems

Energy storage is one of the most important determinants of how long and far an electric powered unmanned aerial system (UAS) can fly. In the case of multi-rotors, lithium polymer (LiPo) batteries are the most commonly used form of energy storage due to their high energy density and high power output. However, even with the most advanced LiPo batteries, electric UAS mission plans are often dictated by the flight time of the vehicle. In most scenarios, this is generally limited to upwards of 30 minutes before battery protection measures activate and the craft must land. This paper addresses this problem by proposing a method of system identification and modeling of energy storage in electric UAS’ to allow for energy storage to be used more effectively in planning autonomous missions. To achieve this, a set of experiments were designed to measure the energy consumption of a mid-size UAS multi-rotor. Several different flight maneuvers were considered: different lateral velocities, climbing, and hovering. The goal of this paper was to create a set of data that allowed each flight maneuver to be characterized with a certain rate of energy usage. Experimentation results demonstrate the feasibility and robustness of the proposed approach. Future work includes the development of mission planning algorithms that provide realistic estimates of possible mission flight times given a certain set of flight parameters.





My graduate research area is in the field of controls and robotics. I’ve had experience in research and design in small electric vehicles and UAVs. In the future, I intend to move towards the field of humanoid robotics such as exoskeletons and standalone humanoid robots.


Figure 1: 3Drobotics quadcopter for research



Figure 2: Preliminary design for lower leg rehabilitative exoskeleton



  1. Carrington, K. Chang, H. Mentis, and A. Hurst, “But I don’t take steps”: Examining the Inaccessibility of Fitness devices for Wheelchair Athletes. ACM SIGACCESS Conference on Computers and Accessibility. ASSETS 2015
  2. D. Clark, K. Chang, and R. S. Gejji, “A Note on Exploring the Stability of the Pupil Light Reflex”, LPS-RP Technical Report, August 2013.
  3. Chang, “Exploring the Stability of the Pupil Light Reflex”, Laboratory for Physical Sciences (LPS) Research Seminar, August 2013 [Invited Talk].
  4. D. Clark, K. Chang, R. S. Gejji, and A. A. Ross, “Methodological Insights to Exploring the Stability of General and Photorealistic Models of the Pupil Light Reflex”, Joint Mathematics Meetings (JMM), January 2014 [Oral Preliminary Report].
  5. Chang, “The Easy Make Oven: Tabletop Scanner”, Summer Undergraduate Research Festival (SURF UMBC), August 2013. [Poster Presentation]
%d bloggers like this: