Comparison of gradient methods for gain tuning of a PD controller applied on a quadrotor system

Jinho Kim; Stephen A. Wilkerson; S. Andrew Gadsden

doi:10.1117/12.2223329

Translator Disclaimer

13 May 2016 Comparison of gradient methods for gain tuning of a PD controller applied on a quadrotor system

Jinho Kim, Stephen A. Wilkerson, S. Andrew Gadsden

Proceedings Volume 9837, Unmanned Systems Technology XVIII; 98370V (2016) https://doi.org/10.1117/12.2223329
Event: SPIE Defense + Security, 2016, Baltimore, Maryland, United States

Abstract

Many mechanical and electrical systems have utilized the proportional-integral-derivative (PID) control strategy. The concept of PID control is a classical approach but it is easy to implement and yields a very good tracking performance. Unmanned aerial vehicles (UAVs) are currently experiencing a signiﬁcant growth in popularity. Due to the advantages of PID controllers, UAVs are implementing PID controllers for improved stability and performance. An important consideration for the system is the selection of PID gain values in order to achieve a safe ﬂight and successful mission. There are a number of diﬀerent algorithms that can be used for real-time tuning of gains. This paper presents two algorithms for gain tuning, and are based on the method of steepest descent and Newton’s minimization of an objective function. This paper compares the results of applying these two gain tuning algorithms in conjunction with a PD controller on a quadrotor system.

Citation Download Citation

Jinho Kim, Stephen A. Wilkerson, and S. Andrew Gadsden "Comparison of gradient methods for gain tuning of a PD controller applied on a quadrotor system", Proc. SPIE 9837, Unmanned Systems Technology XVIII, 98370V (13 May 2016); https://doi.org/10.1117/12.2223329

ACCESS THE FULL ARTICLE