Abstract:Industrial robots are widely used for repetitive, humanly unmanageable, and hazardous tasks. Hence, an improvement in the production efficiency of industrial robot manipulators is of prime concern. This can be achieved through machine vision and path planning techniques with a focus on localization and shortest path calculation. In particular, this is important for manufacturing and bottle filling industries which extensively use robotic manipulators to place/displace bottles during production and post refill placements. This is even more challenging when soft, fragile, or opaque objects have to be detected, since it is significantly difficult for robot vision to focus on their indistinguishable features. To this end, we present an ensemble robot framework with a stereo vision system for tracking colored objects which are sensed using blob analysis. An ensemble robotic framework with neural networks is proposed for predicting and thereby overcoming the inbuilt geometric error present in stereo vision systems. Moreover, we have simplified 2-D correspondence problem to 1-D by using a non-rectified stereo camera model and object tracking by applying the triangulation technique in 3D stereo vision coordinate system (SVCS). Subsequently, the SVCS is transformed into robot stereo vision coordinate system for tracking the object centroid by using an RGB marker placed on the object. Finally, in the learning model we have combined color region tracking with machine learning to achieve high accuracy. The outcomes are in accordance with the designed model and successfully achieve path prediction with up to 91.8% accuracy.

Series Robot Tracking Detection System Based on Stereo Vision

Multi-modal 3D Human Tracking for Robots in Complex Environment with Siamese Point-Video Transformer

Beyond Traditional Driving Scenes: A Robotic-Centric Paradigm for 2D+3D Human Tracking Using Siamese Transformer Network

Human-Computer Interaction System Based on Binocular Stereo Vision

Stereo Vision Based SLAM in Outdoor Environments

Stereo Vision Based SLAM Using Rao-Blackwellised Particle Filter

Accurate and Real-Time 3-D Tracking for the Following Robots by Fusing Vision and Ultrasonar Information

Object Tracking and Positioning Based on Stereo Vision

A Vision-Based Path Planning and Object Tracking Framework for 6-DOF Robotic Manipulator

Binocular stereovision omnidirectional motion handling robot

Multi-object Detection and Tracking by Stereo Vision

A Two-Stage Monocular Vision Detection Method for 6D Pose Estimation in Multi-Heterogeneous Robot Systems

Stereo Vision Based Robot for Remote Monitoring with VR Support

Linear CCD Based Optical Tracking Using Stereo Correspondence Algorithm

UNCALIBRATED 3D ROBOTIC VISUAL TRACKING BASED ON STEREO VISION

A Stereo Vision SLAM with Moving Vehicles Tracking in Outdoor Environment

Robot Manipulator Curve Tracking in Unknown Plane Based on Stereo Vision

A Robust Multi-Object Tracking SLAM System in Dynamic Scenes for Stereo and RGB-D Cameras

Robust Perception-based Visual Simultaneous Localization and Tracking in Dynamic Environments

Method for real-time stereo detection based on multiple features

Deep learning framework for robot for person detection and tracking