UAV Heading Control in Windy and Turbulent Conditions Using Reinforcement Learning
Journal Title: IOSR Journals (IOSR Journal of Electrical and Electronics Engineering) - Year 2018, Vol 13, Issue 1
Abstract
Due to the high non-linearity and coupling of a system model in an Unmanned Aerial Vehicle, UAV, the control of the heading has been a challenging task especially under windy and turbulent conditions. In this paper an online adaptive method using reinforcement learning is proposed to counter the effects of wind disturbances. The heading controller is designed in Matlab/Simulink for controlling a UAV in an X-Plane test platform. Through the X-Plane test platform, the performance of the designed controller is shown using real time simulations under different cross wind conditions. The performance of the proposed method is compared to that of a well tuned PID controller. The results show that the proposed method performs better in tracking a given heading angle under windy conditions.
Authors and Affiliations
Kimathi S, Kang’ethe S. , Kihato P.
Keywords
Related Articles
- EP ID EP388663
- DOI 10.9790/1676-1301032329.
- Views 156
- Downloads 0
Malware Analysis and Mitigation in Information Preservation
Malware, also known as malicious software affects the user’s computer system or mobile devices by exploiting the system’s vulnerabilities. It is the major threat to the security of information in the computer systems. So...
Solar PV/Battery System Analysis for Faculty of Engineering Building, Rivers State University, Port Harcourt, Nigeria.
The incessant power failure in the Faculty of Engineering main building, Rivers State University, Port Harcourt, Nigeria, has made the faculty and departmental offices to resort to private generators that cause air and n...
Performance Analysis of Dstatcom in Live Distribution System for Load Compensation
Power Quality issues are the major concern in the low voltage distribution system. The paper suggests a shunt connected custom power device (DSTATCOM) to provide load compensation in the 3-phase, 4- wire network. The per...
Village load profiling and forecasting model
This paper describes building up of a model for computing the load forecasts as well as producing load profiles of a selected village and comparing it with national load profile. The main requirement before developing th...
Optimal Reactive Power Compensation in Transmission Networks By TCSC Using Particle Swarm Optimization
Increased demand for electrical energy and free market economies for electricity exchange, have pushed power suppliers to pay a great attention to quality and cost of the latter, especially in transmission networks. To r...