At present,most cars are based on ultrasonic and photoelectric sensors to achieve obstacle avoidance and tracking.There are large data acquisition and transmission delays,and the overall work is unstable.We have designed an object tracking car based on image recognition,which can achieve real-time tracking of colored objects and maintain a certain distance.The trolley uses OpenMV(Open Machine Vision)to obtain image information,and transmits the processed image information to the STM32 controller through serial communication.The controller parses and processes the sent data packets.The user uses the human-computer interaction interface to view the image information and adjust the current PWM value and PID control error of the trolley motor.The menu interface switch and parameter adjustment are realized by pressing keys and dialing switches.The design shows that the car can accurately identify color blocks under appropriate lighting conditions,and control the motor to maintain a specific distance,which can achieve real-time tracking of colored objects.
NETL
NSTL
万方数据
