Development of Navigation System on Humanoid Robot Erisa Using Compass Sensor

The ERISA humanoid robot is designed to display stable and directed dance movements in the Indonesian Dance Robot Contest (KRSTI). However, the main challenge faced is the lack of a navigation system that is able to maintain the robot's direction of movement accurately. This study aims to develop a navigation system based on the MPU9250 compass sensor inspired by the STM32F407VGT6 microcontroller. This system is equipped with a PID control algorithm to automatically correct the direction of movement based on heading data obtained in real-time. To improve the accuracy of the sensor data, a Kalman Filter is applied to reduce noise. The designer's research methods include mechanics, hardware development, control algorithm programming, and system testing in various trajectory scenarios, both without interference and in disturbed magnetic field conditions. The test results show that the system is able to maintain direction with a deviation of less than $\pm 1.5^{\circ}$ under normal conditions and a maximum of $\pm 3^{\circ}$ under disturbed conditions. The system also responds to corrections in less than 1 second. In conclusion, this navigation system is able to improve the stability and precision of the ERISA robot's movement, and is suitable for use in robot choreography-based competitions.