Parameter estimation survey for multi-joint robot dynamic calibration case study

Zhang, Shaolin; Wang, Shuo; Jing, Fengshui; Tan, Min

doi:10.1007/s11432-018-9726-3

Cited by 19 publications

(6 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since most collaborative robots rely on geared drives where actuation performance is significantly influenced by frictional effects, it is important to model and identify robots’ joint friction [17, 18]. Typical friction models describe the joint friction torque as a nonlinear and discontinuous function that generally contains components of Coulomb friction, viscous friction, and Stribeck friction [3, 8, 19, 20]. The real values of friction model parameters may be different from their theoretical value due to wear, aging and deformation of the joints.…”

Section: Introductionmentioning

confidence: 99%

An improved iterative approach with a comprehensive friction model for identifying dynamic parameters of collaborative robots

Li,

Wei,

Liu

et al. 2024

Robotica

View full text Add to dashboard Cite

Collaborative robots are becoming intelligent assistants of human in industrial settings and daily lives. Dynamic model identification is an active topic for collaborative robots because it can provide effective ways to achieve precise control, fast collision detection and smooth lead-through programming. In this research, an improved iterative approach with a comprehensive friction model for dynamic model identification is proposed for collaborative robots when the joint velocity, temperature and load torque effects are considered. Experiments are conducted on the AUBO I5 collaborative robots. Two other existing identification algorithms are adopted to make comparison with the proposed approach. It is verified that the average error of the proposed I-IRLS algorithm is reduced by over 14% than that of the classical IRLS algorithm. The proposed I-IRLS method can be widely used in various application scenarios of collaborative robots.

show abstract

Section: Introductionmentioning

confidence: 99%

An improved iterative approach with a comprehensive friction model for identifying dynamic parameters of collaborative robots

Li,

Wei,

Liu

et al. 2024

Robotica

View full text Add to dashboard Cite

show abstract

“…Menq et al (1989) and Menq and Borm (1991) first proposed the concept of an observation index to evaluate and optimize the error measurement configuration of an open-loop planar mechanism and PUMA robot. Subsequently, based on the singular value of the error transfer matrix, scholars have proposed five observation indicators: O 1 , O 2 , O 3 , O 4 and O 5 (Menq and Borm, 1991; Nahvi and Hollerbach, 1996; Wang et al , 2017; Zhang et al , 2019; Sun and Hollerbach, 2008), and their effectiveness has also been widely investigated. The O 1 observation index is the most effective when the variation of the minimum identification parameters is taken as the objective (Menq et al , 1989; Menq and Borm, 1991; Qiao et al , 2021; Daney et al , 2005).…”

Section: Introductionmentioning

confidence: 99%

Error identification and compensation of 1T2R parallel power head based on trajectory optimization and principal component analysis

Cui

Song

et al. 2023

View full text Add to dashboard Cite

Purpose The purpose of this paper is to propose a method for selecting the position and attitude trajectory of error measurement to improve the kinematic calibration efficiency of a one translational and two rotational (1T2R) parallel power head and to improve the error compensation effect by improving the properties of the error identification matrix. Design/methodology/approach First, a general mapping model between the endpoint synthesis error is established and each geometric error source. Second, a model for optimizing the position and attitude trajectory of error measurement based on sensitivity analysis results is proposed, providing a basis for optimizing the error measurement trajectory of the mechanism in the working space. Finally, distance error measurement information and principal component analysis (PCA) ideas are used to construct an error identification matrix. The robustness and compensation effect of the identification algorithm were verified by simulation and through experiments. Findings Through sensitivity analysis, it is found that the distribution of the sensitivity coefficient of each error source in the plane of the workspace can approximately represent its distribution in the workspace, and when the end of the mechanism moves in a circle with a large nutation angle, the comprehensive influence coefficient of each sensitivity is the largest. Residual analysis shows that the robustness of the identification algorithm with the idea of PCA is improved. Through experiments, it is found that the compensation effect is improved. Originality/value A model for optimizing the position and attitude trajectory of error measurement is proposed, which can effectively improve the error measurement efficiency of the 1T2R parallel mechanism. In addition, the PCA idea is introduced. A least-squares PCA error identification algorithm that improves the robustness of the identification algorithm by improving the property of the identification matrix is proposed, and the compensation effect is improved. This method has been verified by experiments on 1T2R parallel mechanism and can be extended to other similar parallel mechanisms.

show abstract

“…Extended Kalman filter (EKF), based on the first-order Taylor expansion of KF, provides higher accuracy estimation results for nonlinear systems. In terms of its low computation complexity and ease of implementation, it is regarded as the powerful tool for online parameter estimation [21]. Additionally, it has been widely applied in guidance, navigation, and control of robots and spacecraft in unknown environments [22][23][24][25].…”

Section: Introductionmentioning

confidence: 99%

Continuous-discrete extended Kalman filter based parameter identification method for space robots in postcapture

Zhang,

Shi,

Yang

et al. 2023

Preprint

View full text Add to dashboard Cite

Space robots have become increasingly important in on-orbit missions, especially for capturing noncooperative targets. However, a major challenge is that the inertial parameters of these targets are often unknown, but crucial for post-capture tasks. This paper proposes continuous-discrete extended Kalman filter (CD-EKF) based identification methods that rely solely on noisy measurements of the manipulator’s rotation angle, the base’s attitude angle, and its position. The methods exploit the conservation of momentum or the dynamics of space robots to formulate the identification equations and construct the filter. In addition, an analytical solution for computing the Jacobian matrix of the dynamic response of a space robot is derived, and sparse matrix multiplication is used to reduce the computation time. The effectiveness and efficiency of the proposed methods are evaluated through numerical simulations using 2D and 3D models, and Monte Carlo simulations are performed to analyze robustness, noise effects, and initial states. The simulation results confirm the effectiveness of the proposed methods and demonstrate their ability to handle noise and uncertainty.

show abstract

Parameter estimation survey for multi-joint robot dynamic calibration case study

Cited by 19 publications

References 38 publications

An improved iterative approach with a comprehensive friction model for identifying dynamic parameters of collaborative robots

An improved iterative approach with a comprehensive friction model for identifying dynamic parameters of collaborative robots

Error identification and compensation of 1T2R parallel power head based on trajectory optimization and principal component analysis

Continuous-discrete extended Kalman filter based parameter identification method for space robots in postcapture

Contact Info

Product

Resources

About