Dynamic Error Reduction via Continuous Robot Control Using the Neural Network Technique

Глазков, В. П.; Daurov, Stanislav; Львов, А. А.; Askarova, Adel; Kalikhman, D. M.

doi:10.1007/978-3-030-65283-8_15

Cited by 3 publications

(2 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Various error sources influence robot accuracy during operation, including, geometric defects [2], non-geometric uncertainty [3], kinematic and non-kinematic errors [4,5], dynamic errors [6], drive system backlash [7,8], thermal effect [9][10][11] and system vibrations [12,13].…”

Section: Introductionmentioning

confidence: 99%

Tool Positioning Error Minimization during Robotic Milling Based on the Genetic Algorithm Technique

Lounici

Ouali

Osmani

2022

AMM

View full text Add to dashboard Cite

Compared to CNC machines, robotic milling has performance limitations such as accuracy and quality. The main source of the robot’s inaccuracy during machining is the flexibility of its parts (body or joints). This error disturbs the movement of the end effector, affecting the part’s surface finish. In order to improve the robot’s accuracy and minimize the positioning error of the end effector during the milling operation, this paper presents, first, a method based on the elasto-static model to predict the Cartesian deflection of the end effector of a three DOF redundant planar robot, and second, optimization techniques with original objective functions based on the single and multi-objective genetic algorithm, which will be presented and compared. The programming of the two methods and the results of the study will be done using MATLAB software. The analysis of simulation results of the two optimization techniques GA and MOGA revealed that the tool configuration and cutting parameters used for robotic milling have a direct influence on the robot's path accuracy and milling performance. Whereas for a φ0=69.6, φf=72.43 the maximum tool deviation in its path is Δxmax ≈ |0.125| mm with a maximum roughness profile height Ra = 1600 μm. While the positioning error is said to be minimal Δxmin ≈ |0.025| when φ0= -38.67, φf = -35.92, and the roughness Ra= 25 μm.

show abstract

Section: Introductionmentioning

confidence: 99%

Tool Positioning Error Minimization during Robotic Milling Based on the Genetic Algorithm Technique

Lounici

Ouali

Osmani

2022

AMM

View full text Add to dashboard Cite

show abstract

“…These indicators, as well as other criteria, are thoroughly documented in the standard [1]. Various error sources influence robot accuracy during operation, including geometric defects [2], non-geometric uncertainty [3], kinematic and non-kinematic errors [4,5], dynamic errors [6], drive system backlash [7,8], thermal effect [9][10][11], and system vibrations [12,13].…”

Section: Introductionmentioning

confidence: 99%

Cutting Forces Impact on the Spindle Path during Robotic Milling

Lounici

Ouali

Osmani

2022

AMM

View full text Add to dashboard Cite

Offline programming is a critical step in the implementation of various robotic tasks such as pick-and-place, welding, cutting, and milling. This paper describes a simulation study that analyses the accuracy of the robot's path tracking, during tasks that require the robot tool to interact with the environment, while considering the current operating conditions. To accurately determine the actual position of the Tool Center Point (TCP) and the associated orientation of the end effector, the study will first establish a robot model that takes into account the elasto-static behavior during the milling process that generates significant contact forces on the end effector. Then, an offline simulation tool is developed within the SolidWorks® CAD environment. The analysis of simulation results from multiple scenarios revealed that the tool/material contact forces were the main source of the robot's deviation from its nominal trajectories. Moreover, the range of positioning errors varies according to the architecture of the robot and the workpiece emplacement. Depending on the working conditions, the tool deflection ranges from 0.1 mm to 0.75 mm in the or cutting directions and increases as one moves away from the reference frame, while the Cartesian orientation deviation is negligible (less than 1°).

show abstract

Ensuring the Robustness of Modern Mechatronic Systems Using Artificial Intelligence Methods

Kalikhman

Deputatova

Львов

et al. 2023

Studies in Systems, Decision and Control

View full text Add to dashboard Cite

Dynamic Error Reduction via Continuous Robot Control Using the Neural Network Technique

Cited by 3 publications

References 16 publications

Tool Positioning Error Minimization during Robotic Milling Based on the Genetic Algorithm Technique

Tool Positioning Error Minimization during Robotic Milling Based on the Genetic Algorithm Technique

Cutting Forces Impact on the Spindle Path during Robotic Milling

Ensuring the Robustness of Modern Mechatronic Systems Using Artificial Intelligence Methods

Contact Info

Product

Resources

About