Energy Optimization in Single and Dual UAV Collaborative Aerial Transportation

Abstract

An Unmanned Aerial Vehicle (UAV) carrying a payload can be used for multiple operations, which could be search and rescue, fire extinguishing etc. in environments inaccessible or unfavorable for humans. Recently UAVs have been used to carry organs for transplants, medical samples and even coffee. In case of medical samples, or organs for transplants, it is important to reduce journey time, whereas in other cases energy saving is important. It is therefore useful to be able to switch between energy saving or time saving while doing a UAV based payload transportation. Aerial transportation with multi-rotors is constrained by the available energy in the batteries. This thesis focuses on approaches to optimize the energy consumption or time spent in journey in single and dual UAV aerial transportation while also considering energy distribution in dual-UAV aerial transportation. This thesis deals with the application of Real Time Dynamic programming (RTDP) in velocity optimization for single and dual-UAV aerial transportation. RTDP is applied for the first time for the optimization of the motion in single and dual UAV aerial transportation. At first, the RTDP algorithm was developed for single UAV payload transportation where multi-criteria optimization was performed via minimizing a cost function. The cost function included a factor _ which can be used to switch the priority between energy consumption and time spent. Mathematical model of single UAV payload system was developed in MATLAB/Simulink environment, and Gazebo model was developed for the same UAV and was used in software in the loop simulations for testing of RTDP algorithm. Hardware experiments were also performed to test the applicability of the proposed algorithm. After that energy distribution between the two UAVs jointly transporting a payload was studied since power failure in either UAV can fail the whole operation. The dual-UAV collaborative transport strategy for distributing the lifting load between the two UAVs to distribute power consumption was developed and tested in Simulink and hardware experiments. The strategy successfully distributed the lift force for load with a ratio 62=38% and showed 8% energy distribution in experiments. In simulations 24% energy distribution was achieved. RTDP algorithm was then developed for dual-UAV-payload transportation. At first, the model of the dual-UAVpayload system was defined. Based on the definition of the model, the cost function calculation was defined while considering the possibility of energy or load distribution during the collaborative transportation. The effects of the definition of cost function on algorithm complexity were discussed. The developed algorithm was further tested in the simulink based numerical model, the Software in the loop Gazebo simulations and then hardware experiments were performed to show the applicability and effectiveness of the algorithm.

Date of Award	Sep 2019
Original language	American English