Calibration and uncertainty analysis of a combined tracking-based vision measurement system using Monte Carlo simulation

Recently, visual sensing measurement and its application in industrial robot operations have been widely researched, promoting the development of instrumentation and automation. This study proposes a combined vision sensor system for robot grasping, focusing on combined sensor system calibration and bracket pose measurements. The system configuration and working strategy of the combined vision system are introduced. Thereafter, the calibration of the combined vision coordinate systems is presented, wherein a global vision system (GVS) acts as the external measuring equipment for accurately calibrating the local vision. Furthermore, a pose estimation method using a local vision system (LVS) is proposed, including morphology-based image enhancement and principal component analysis (PCA)-based corner recognition methods. Verification experiments, including combined calibration and bracket pose measurements, were performed to validate the effectiveness and accuracy of the proposed combined vision measurement strategy. The results demonstrated that the proposed system applies to industrial robot grasping of brackets. In addition, the proposed robot-sensor calibration method improves calibration accuracy. Finally, the proposed corner detection method is effective and accurate for different bracket detection applications. This study provides a system that improves robot grasping results by considering key factors, such as vision measurement accuracy, and calibration methods.

Section: Related Workmentioning

confidence: 99%

Calibration and pose measurement of a combined vision sensor system for industrial robot grasping of brackets

Jiang,

Deng,

Yin

et al. 2024

“…The key procedure for visual measurement is the motion extraction of image sequence. There are several methods for motion extraction [23][24][25]. In this paper, the phase-based motion extraction method suggested by Peng et al [25] was used to extract the displacement time histories of the measurement points in the video.…”

Section: Phase-based Visual Measurementmentioning

confidence: 99%

Identification of vehicle axle loads based on visual measurement

Liu¹,

Wang²,

Deng³

et al. 2022

Monitoring vehicle axle loads is very important for preventing infrastructure degradation and traffic accidents. However, developing a cost-effective, flexible, and accurate axle load identification technology still remains challenging. In this study, a vehicle axle load identification method based on visual measurement is proposed. The principle of the method is that each axle load of the vehicle is determined based on the ratio of vehicular sprung mass to unsprung mass and the centroid position of the sprung mass, which can be predicted from the visually captured free decaying vibrations of the vehicle as it passes over a speed bump. To be more specific, the vibration responses of a vehicle passing over a speed bump is firstly measured by a camera, and the vehicle system matrix, which is composed of the mass matrix, stiffness matrix and damping matrix, can be obtained from the vehicle responses. The axle load can then be determined based on the element ratios in the system matrix of the vehicle whose unsprung masses are known. The performance of the proposed method is evaluated through numerical simulations and field tests. Results show that the identified axle loads agree well with the corresponding true values, proving the effectiveness and accuracy of the proposed method. This study has demonstrated a novel application of the computer vision technology to identify the axle loads of moving vehicles. The proposed method does not require the installation of sensors on the roadway or on the vehicle, making it a promising alternative for traditional weigh-in-motion systems.

“…The linear profile sensor (LPS) has occupied a significant status in the field of 3D measurement for its high collection frequency, simplicity, and cost-effectiveness [15,[24][25][26][27][28][29][30][31]. Unlike RGB-D cameras and binocular vision instruments, the LPS employs a high-intensity, monochromatic light source, which renders it less susceptible to interference from ambient light, surface textures or colors of objects.…”

Section: Introductionmentioning

confidence: 99%

“…Several researchers have proposed solutions to address this challenge. For instance, Jiang et al [29] and Liu et al [30] explored the measurement system incorporating LPS and rotating workbenches. Li et al [31] established a scanning measurement system by combining a linear laser with a galvanometer.…”

Section: Introductionmentioning

confidence: 99%

A density-controllable 3D measurement system based on linear profile sensor and universal manipulator

Zhao,

Zhang,

2024

In order to offer a simple and efficient solution for obtaining point clouds, a modular 3D measurement system that incorporates an LPS mounted on a universal manipulator is developed. Unlike the requirement for electronic connections between the LPS and the robot to achieve hard synchronization, we employ a high-low-frequency collaborative strategy for soft synchronization, thereby eliminating the need for complex hardware synchronization modules. Since soft synchronization lacks the high-speed synchronous data acquisition of hard synchronization, we have also proposed an interpolation method on low density data to generate uniform higher-density point clouds, thus enabling a density-controllable point clouds acquisition approach. Experimental results demonstrate the effectiveness and versatility of the system.