In this thesis, we designed, built, and tested a prototype of a line laser depth scanner for a robot, based on the principle of optical triangulation. The project encompassed the selection and integration of hardware components (laser, camera, mounts, computer), the development of software in Python using OpenCV, the implementation of system calibration, testing under controlled conditions, and the analysis of accuracy and possibilities for improvement. The goal was to create a cost-effective, modular system suitable for basic technical applications.
In the first chapter, we described the basic principles of laser operation, their types and characteristics, with an emphasis on diode lasers used in depth sensing. We addressed the safety aspects of laser use and the principle of optical triangulation as the foundation of the line laser scanner. We provided a detailed description of the scanner components (laser, camera, optical elements, computing unit, mechanical components) and presented examples of their applications in industry, science, biometrics, robotics, and medicine. Finally, we defined the technical, functional, and safety requirements for the development of the prototype.
In the second chapter, we presented the entire process of designing, developing, and building the prototype. We described the selection of components according to technical requirements, the mechanical assembly of the laser module and camera, and the integration of electronic components. We presented the development of software in Python using the OpenCV library for image capture and processing, the algorithm for processing captured images, and the calculation of depth data using triangulation. We carried out error minimization based calibration of the sensor parameters and verified the system’s operation in a test environment. Based on the tests, we analysed the results and errors (average error of 1,2 mm) and evaluated the stability and repeatability of the system’s operation.
In the fourth chapter, we summarized the key findings—the prototype achieved the planned accuracy and met the technical requirements, while also being modular and suitable for further upgrades. We highlighted the system’s limitations, such as sensitivity to reflective surfaces and a limited scanning angle, and proposed improvements such as using higher-quality optics, integrating a moving mechanism, and implementing hardware acceleration for image processing.
|
