Design and realization of an energy efficient vision-based autonomous target tracking drone using LabVIEW

UAVs based surveillance system adopted in various countries for undergoing rescue services in mountainous terrains during natural calamities, highly vehicular traffic regions and in fire rescue operations. The static altitude maintenance requires frequent recharging of batteries with poor flight time. The Centralized Altitude Scheduling (CAS) and Localized Altitude Scheduling (LAS) Algorithms cannot produce the optimum solution even for small instances because the altitude of the drone is calculated in prior to the surveillance action, it may be suitable for the static targets but produces inefficient results for the mobile target objects. The options and features available to the users in existing autonomous surveillance system are minimum and cannot be used in places other than the application it is designed for. Here, this paper focuses on vision based altitude adjustment method, a solution for the above mentioned problem which introduces the ability of the drones to automatically adjust its altitude by estimating the size of the target on which surveillance is being performed. Automatic Altitude Adjustment (AAA) is implemented on the basis of Mean Shift Algorithm by formulating the probability distribution function (PDF) for the target specified inside Region of interest ROI. The different cases regarding the autonomous navigation of drones and the vision based altitude scheduling approach are carried out using graphical programming in LabVlEW software. The results have been compared to the existing drone based surveillance systems and the eventually differences have been discussed.