Depth Map Reconstruction Method in Control Problems for Robots and Mechatronic Systems

Abstract

In modern robotic and mechatronic systems, technologies are in demand that makes it possible to build an optimal trajectory of movement of their actuators. Such technologies are formed by combining navigation methods and building a 3-D map of the surrounding space based on vision systems and are successfully used in robotics and mechatronics. But there is a problem, consisting of a decrease in the accuracy of planning the trajectory of movement, caused by incorrect sections on the map (depth map) due to incorrect determination of the distance to objects. Such defects appear as a result of poor lighting, specular or fine-grained surfaces of objects. This leads to the impossibility of obtaining reliable information about the depth. As a result, the effect of increasing the boundaries of objects (obstacles) appears, and the overlapping of objects makes it impossible to distinguish one object from another. This problem can be solved using image reconstruction methods. The article presents an approach based on a modified algorithm for searching for similar blocks using the concept of quaternions and anisotropic gradient. The analysis of the research results shows that the proposed method allows you to correctly restore the boundaries of objects on the depth map image when reconstructing 3-D scenes, which contributes to an increase in the accuracy of planning the trajectory of motion of the actuators robotic and mechatronic systems.