Our group adopts machine learning (ML) techniques that are designed to predict material removal rates, achievable surface roughness, and create virtual samples that are essential for surface defect detection and prediction [1]. We have identified techniques that employ easily obtainable input data, such as cutting forces and vibration signals, to predict milling surface roughness using a fuzzy broad learning system algorithm with self-learning capabilities [2]. We also have the echo state broad learning system algorithm for virtual sample creation [3] and the BESLS for material removal predictions during internal polishing [4].

We also develop intelligent tool path planning algorithms involving high-order calculations [5] to accommodate machine tool stiffness [6], orientation, and positioning for material removal [7,8]. Additionally, we have developed solutions to achieve time-optimized tool path planning to maximize efficiency, which will be particularly useful for the post-processing of surfaces with complex curvatures [9].

Publications

[1] T. Niu, et al., A generalized well neural network for surface defect segmentation in Optical Communication Devices via Template-Testing comparison, Computers in Industry, 151 (2023) 103978.

[2] W. Tian, et al., Prediction of surface roughness using fuzzy broad learning system based on feature selection, Journal of Manufacturing Systems, 64 (2022) 508–517.

[3] W. Tian, et al., Interpolation-based virtual sample generation for surface roughness prediction, Journal of Intelligent Manufacturing, 35 (2024) 343–353.

[4] J. Zhang, et al., Material removal rate prediction based on broad echo state learning system for magnetically driven internal finishing, IEEE Transactions on Industrial Informatics, 19 (2023) 6295–6304.

[5] L. Lu, et al., High-order joint-smooth trajectory planning method considering tool-orientation constraints and singularity avoidance for robot surface machining, Journal of Manufacturing Processes, 80 (2022) 789–804.

[6] P. Xu, et al., Stiffness modeling of an industrial robot with a gravity compensator considering link weights, Mech. Mach. Theory, 161 (2021) 104331.

[7] L. Lu, et al., Tool path optimization for robotic surface machining by using sampling-based motion planning algorithms, Journal of Manufacturing Science and Engineering, 143 (2020) 011002.

[8] L. Lu, et al., Joint-smooth Toolpath Planning by Optimized Differential Vector for Robot Surface Machining Considering the Tool Orientation Constraints, IEEE/ASME Transactions on Mechatronics, 27 (2022) 2301–2311.

[9] L. Lu, et al., Time-optimal tool motion planning with tool-tip kinematic constraints for robotic machining of sculptured surfaces, Robotics and Computer-Integrated Manufacturing, 65 (2020) 101969.