A New Technique for Improving the Precision of Dual-Axis Tetrahedron-Based Sensor With Switching Between PID and ANN

This paper proposes a new technique for a dual-axis light tracking system using a combination of Proportional Integral Derivative (PID) and Artificial Neural Network (ANN) with a tetrahedron-based sensor. A tetrahedron-based sensor can detect the position of light using three photoresistor components arranged on a tetrahedron plane. The proposed technique combines PID and ANN controllers using a switching method that allows the algorithm to switch between PID and ANN based on the occurring errors. The updating of PID controller gains is done offline. Additionally, in the proposed tracking system, a new mechanism called dual-axis in dual-frame is introduced. This mechanism involves the use of two frames moving in parallel along opposite axes. Experimental results collected the angular movement data and the decrease in error values. Numerical analysis showed that the proposed prototype of the PID-ANN tracking system demonstrates better response time than previous prototypes mentioned in the literature. Moreover, this system also exhibits low error rates and improved data distribution. Therefore, the PID-ANN tracking system with the switching process enhances the performance of responsive, accurate, and precise light tracking.