Develoment of the robot remote control system

Abstract

In modern technologies across industry, military affairs, medicine, science, and other fields, robot, equipped with machine vision which consist of mechanical manipulators and their control systems are widely used. Тhe aim of this work is to expand the functional capabilities of the control system of the robot to safely grasp an object of arbitrary shape. To achieve this goal, a hardware–software subsystem, has been developed. The software and hardware complex consists of such main parts as the user, the web client, the devices part, functional block. Appropriate third-party libraries are used for the program code. The hardware part of the system was developed using a Raspberry Pi microcomputer, a Stereo Pi board, a board for connecting eight servomotors (Adafruit PCA9865), and Raspberry Pi camera modules. The cameras are connected to the microcomputer via a wired connection. The development of the system’s software included creating the manipulator’s software with machine vision cameras, the microcomputer, and a web application, taking into account established control systems. The Python language is used throughout the developed system. For the web application, the WebStorm development environment from JetBrains and the Flask framework were chosen. The functional part of the developed system performs tasks such as moderation, collective administration, user feedback, realtime programming of the microcomputer, and connecting to a local network or the Internet. A distinctive feature of this functional part is that it enables, remotely via the web application, the adjustment and fine-tuning servomotor precision. It also allows for the involvement of a team of competent users – including both unauthorized and authorized users, experts, and administrators. Performance testing of the system, under the simultaneous operation of a large group of users (up to a thousand), was conducted using Apache JMeter. In addition, the system’s software was tested; both functional and module testing were carried out on the developed interface. The study of the speed of the system during the simultaneous work of a large group of users (up to a thousand) was carried out using the Apache JMeter program. During functional testing, appropriate messages, redirections, manipulator actions, dialogues, and user interactions were successfully implemented. For module testing of the web application, the Pytest framework was used.