A generalized algorithm for tuning UAS flight controllers

Proportional, Integral and Derivative (PID) controllers are among the most commonly used control systems throughout industry, and there is an increasing need to tune such controllers effectively and rapidly, especially in varying dynamic conditions. Here we present two versions of a generalized, iterative method for tuning PID controllers: Iterative Root Mean Square Optimization (iRMSE) and Iterative Weighted Root Mean Square Optimization (iWRMSE). The two methods are validated in Matlab and in a virtual environment, as well as in field tests with a quadcopter. The performance of our two methods are compared against five popular methods: Zeigler-Nichols, Cohen-Coon, Lambda, Root Mean Square Error (RMSE) and Integral Square Error (ISE). We find that iWRMSE optimization delivers performance that is better than that obtained using all of the other methods, including manual tuning. Both iRMSE and iWRMSE can be used on a wide range of systems. Due to their iterative nature, they are also likely to be more suitable for systems operating in noisy or variable environments.