Poster + Paper
7 August 2023 The algorithm for improving the accuracy of positioning and recognition of objects on a machine vision system used for direct control of a joint manipulator on flexible cells
Author Affiliations +
Conference Poster
Abstract
This article considers the construction of an algorithm for automated analysis of the position of objects in a limited space (field, workshop, warehouse) with the possibility of constructing a control action for moving the camera and gluing video data. The method of parallel data analysis allows, in addition to constructing the optimal movement trajectory, to integrate the video subsystem into the manipulator control system, which will also allow assessing the relative position of the operator, cargo and obstacles located in the manipulator service area. The paper presents the structure and description of the construction of a video subsystem consisting of several neuroaccelerators with RISC-V core controllers and a video camera system. The structure of control formation for a group of controllers, united in a single network with a central computer that transmits pre-processed data to the control system, is presented. The presented technique makes it possible to form a system including: controllers, video processing, a computer and a control system that allow real-time processing.
© (2023) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Andrey Alepko, Yury Valukevich, Aleksandr Shaldov, Tagir Abdullin, and Evgenii Semenishchev "The algorithm for improving the accuracy of positioning and recognition of objects on a machine vision system used for direct control of a joint manipulator on flexible cells", Proc. SPIE 12624, Digital Optical Technologies 2023, 126241R (7 August 2023); https://doi.org/10.1117/12.2678222
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Control systems

Cameras

Video

Imaging systems

Machine vision

Kinematics

Sensors

Back to Top