Date of Award
Master of Science (MS)
Modeling Simul & Visual Engineering
Modeling & Simulation Engineering
The objective of this investigation is to explore the use of hand gestures to control semi-autonomous vehicles, such as quadcopters, using realistic, physics based simulations. This involves identifying natural gestures to control basic functions of a vehicle, such as maneuvering and onboard equipment operation, and building simulations using the Unity game engine to investigate preferred use of those gestures. In addition to creating a realistic operating experience, human factors associated with limitations on physical hand motion and information management are also considered in the simulation development process. Testing with external participants using a recreational quadcopter simulation built in Unity was conducted to assess the suitability of the simulation and preferences between a joystick approach and the gesture-based approach. Initial feedback indicated that the simulation represented the actual vehicle performance well and that the joystick is preferred over the gesture-based approach. Improvements in the gesture-based control are documented as additional features in the simulation, such as basic maneuver training and additional vehicle positioning information, are added to assist the user to better learn the gesture-based interface and implementation of active control concepts to interpret and apply vehicle forces and torques. Tests were also conducted with an actual ground vehicle to investigate if knowledge and skill from the simulated environment transfers to a real-life scenario. To assess this, an immersive virtual reality (VR) simulation was built in Unity as a training environment to learn how to control a remote control car using gestures. This was then followed by a control of the actual ground vehicle. Observations and participant feedback indicated that range of hand movement and hand positions transferred well to the actual demonstration. This illustrated that the VR simulation environment provides a suitable learning experience, and an environment from which to assess human performance; thus, also validating the observations from earlier tests. Overall results indicate that the gesture-based approach holds promise given the emergence of new technology, but additional work needs to be pursued. This includes algorithms to process gesture data to provide more stable and precise vehicle commands and training environments to familiarize users with this new interface concept.
"Gesture Based Control of Semi-Autonomous Vehicles"
(2019). Master of Science (MS), Thesis, Modeling Simul & Visual Engineering, Old Dominion University, DOI: 10.25777/9z93-c188