Yapay Zeka Hızlı Bakış & Çıkarımlar
AI InsightIt was clarified that using PID controller and Kalman Filter together have a positive effect to balance the robot on the desired angle.
📌 3 Maddede Bu Makale
Özet
Real-time robots are quite common in our daily life. These robots are working as a part of the process in industry or a medical assistance in hospitals to serve humanity. Designing the robots according to the desired referent and making the given tasks with high accuracy makes them more and more popular in these days. In this work, the designed two-wheeled balancing robots with integrated camera track object autonomously. This work has two important stages. The first stage is about balancing the robot with the angle information taken from IMU sensor and implementation of PID control. IMU sensors create lots of noisy signals because of its natural structures. Kalman filter was used to denoise these noisy signals to have a smooth signal for a better balance control. The second stage is about image processing and objects recognition. This section was completed by using Matlab Image Processing Toolbox which can be used Arduino microcontroller board synchronously. In this section, algorithm infers motion information of objects. Motors were controlled according to motion information of moving objects. In the end, an object tracker self-balance robot was constructed. Balance control of the robot was managed by PID controller and accelerometer signals were denoised by a Kalman Filter. It was clarified that using PID controller and Kalman Filter together have a positive effect to balance the robot on the desired angle.
Abstract
Real-time robots are quite common in our daily life. These robots are working as a part of the process in industry or a medical assistance in hospitals to serve humanity. Designing the robots according to the desired referent and making the given tasks with high accuracy makes them more and more popular in these days. In this work, the designed two-wheeled balancing robots with integrated camera track object autonomously. This work has two important stages. The first stage is about balancing the robot with the angle information taken from IMU sensor and implementation of PID control. IMU sensors create lots of noisy signals because of its natural structures. Kalman filter was used to denoise these noisy signals to have a smooth signal for a better balance control. The second stage is about image processing and objects recognition. This section was completed by using Matlab Image Processing Toolbox which can be used Arduino microcontroller board synchronously. In this section, algorithm infers motion information of objects. Motors were controlled according to motion information of moving objects. In the end, an object tracker self-balance robot was constructed. Balance control of the robot was managed by PID controller and accelerometer signals were denoised by a Kalman Filter. It was clarified that using PID controller and Kalman Filter together have a positive effect to balance the robot on the desired angle.