Prof Yan Jin

Published in the IET Cyber-Physical Systems: Theory and Applications Journal

Within the Industry 4.0 landscape, humans collaborate with cyber and physical elements to form human-cyber-physical systems (HCPS). These environments are increasingly complex and challenging workspaces due to increasing levels of automation and data availability. An effective system design requires suitable frameworks that consider human activities and needs whilst supporting overall system efficacy.

Although several reviews of frameworks for technology were identified, none of these focused on the human in the system (moving towards Industry 5). The critical literature review presented provides a summary of HCPS frameworks, maps the considerations for a human in HCPS, and provides insight for future framework and system development. The challenges, recommendations, and areas for further research are discussed.

This paper was presented at the 2023 3rd International Conference on Mechanical, Aerospace and Automotive Engineering (CMAAE 2023) held in Nanjing, China.

A buffer is an important element in a production line and its allocation influences the throughput and inventory of the line. The buffer allocation problem can be framed as a multi-objective optimisation problem and is often addressed by using meta-heuristic algorithms, such as evolutionary algorithms. However, these algorithms primarily focus on the “genetic evolution” aspect and do not take in to account the impact of the biological “organism development” process, potentially constraining the exploration of the solution space. In this paper, a bio-inspired evolution-development (evo-devo) approach for modelling and optimising buffer allocations is proposed. The organism representing a production line is defined and modelled, and the evolution and development processes of organisms are developed for researching optimised solutions. The method has been validated by a simulation of a buffer allocation optimisation in an unreliable serial production line with multi-objectives, aiming to maximize production throughput and minimize the total buffer size. Results show that the
proposed method can efficiently obtain solutions, while also achieving greater exploration of the solution space than competing evolutionary algorithms such as the Non-Dominated Sorting Genetic Algorithm II. The proposed approach’s functionality means that it could be applied to other areas of generative design of future factories.

Conference Publication by Laura McGirr in June 2022 at the ISR EU2022 Conference in Munich, Germany.

The goal of the research outlined in this paper is to facilitate improved communication and
enable co-ordinated research collaboration across academic researchers and industry in the implementation of HRC installations.

This has been published in the “Engineering” on-line Journal, accessible on ScienceDirect.

Engineering | Journal | ScienceDirect.com by Elsevier

Abstract

Controlling machining deformations resulting from unbalanced stress fields inside structural components is a significant challenge in the manufacturing industry. Prediction of machining deformation fields is fundamental for deformation control and requires numerous iterations to optimize the machining process. Conventional prediction methods such as numerical analysis are tailored to a fixed geometry, making them time-consuming and inefficient for components with various geometries. In this study, a general data-driven model is proposed for predicting machining deformation fields in components with varying geometries and stress fields. This model is based on a geometry-oriented neural operator that incorporates global geometry information into the function space, modeling the relationship between the input function (stress fields) and the output function (deformation fields). Global geometric information is extracted using a graph neural network applied to a geometric graph and embedded into the input and output function space through an encoder-query framework. The proposed model achieved low root-mean-squared errors ranging from 0.001 to 0.016 mm, with maximum prediction errors between 0.003and 0.047 mm across different types of components, including beams and frames. The main contribution of this research is the significant advancement in the application of neural operators to the development of general models for predicting machining deformation. The underlying principles of the proposed model provide an important reference for wider applications related to the control of machining deformation in the context of digital and intelligent manufacturing.

This paper was presented by Caitlin Sands at the 9th International Workshop on New Trends in Medical and Science Robotics (MESROB 2025) Conference in Poitiers, France July 2-4. The paper can be found in the Proceedings accessible via New Trends in Medical and Service Robotics: MESROB 2025 | SpringerLink. The paper itself can also be found via https://link.springer.com/chapter/10.1007/978-3-031-96081-9_4

Abstract. Due to the high demand for ophthalmic surgery and a shortage of surgeons, robot-assisted surgical systems have attracted attention from research communities for their potential to provide superior accuracy, precision, and motion stability, thus reducing the risk of hand tremors associated with manual surgery. However, existing surgical robotic platforms are less adaptable and efficient than traditional manual procedures. Additionally, the design process for these systems is challenging and time-consuming, requiring specialised expertise from multiple stakeholders. To address these challenges, this paper proposes a framework for applying generative design (GD) methods in developing surgical robots for ophthalmic procedures. It aims to generate diverse, feasible solutions, guided by input requirements and system analyses while facilitating continuous improvements to meet future needs. The paper presents the initial implementation of the proposed framework, which includes the development of a library of link geometries and fundamental 1-degree-of-freedom (DoF) mechanisms. These serve as the building blocks for remote centre of motion (RCM) mechanisms, which are critical for minimally invasive surgery (MIS) applications. The paper discusses the future development of the GD framework, aiming to enhance its capabilities for more complex surgical robotic mechanisms. It focuses on refining the framework to address scalability and validation challenges, ultimately improving design efficiency and adaptability for advanced systems

Published in The International Journal of Advanced Manufacturing Technology, accessible via the following links https://link.springer.com/journal/170 and https://doi.org/10.1007/s00170-026-17862-8

Abstract
Tolerance allocation is an important design step for determining robot accuracy and directly affecting manufacturing cost. However, existing methods typically consider dimension synthesis first before tolerances are allocated, which neglects the manufacturability constraints arising from the dependency between part size and achievable tolerance grades. This often leads to costly iterations between design and manufacturing and increasing manufacturing cost. To address this issue, an integrated tolerance allocation and dimensional synthesis method of robot design is proposed for optimizing both positioning accuracy reliability and manufacturing cost. The method simultaneously optimizes joint dimensions and corresponding tolerances by formulating a cost function that captures the relationship between dimensional parameters, robot end-effector accuracy reliability, tolerance-grade rules, and manufacturing cost. Additionally, a matrix-based Monte Carlo simulation (MCS) method is introduced to accelerate evaluation workspace-wide reliability under tolerance uncertainty. NSGA-II multi-objective optimization algorithm is employed to find the Pareto front of the optimal solutions. A case study of a surgical robot is taken to demonstrate the effectiveness of the proposed approach. Results show that the proposed method can reduce 22% of manufacturing cost while achieving better positioning accuracy reliability compared to the traditional tolerance allocation method, and the speed of matrix-based MCS method is improved by 400 times compared to the point-based MCS method.

This paper will included in the Advanced Engineering Informatics Journal, Volume 73, accessible via the following links https://www.sciencedirect.com/science/article/pii/S1474034626002922 and https://doi.org/10.1016/j.aei.2026.104600

Abstract

Robotic design and optimization often require extensive simulation-based iterations, particularly for predicting static deformations under varying postures, external loads, and gravitational conditions. Conventional finite element simulations can achieve high accuracy but are computationally expensive, limiting their use in large-scale design optimization. Data-driven approaches have emerged to accelerate prediction; however, most existing methods are restricted to simplified scenarios, such as fixed geometries with varying poses or fixed poses with varying loads, and typically predict only partial deformation at points rather than the full deformation field. To enable fast and generalizable deformation prediction, a new neural operator architecture named Robot-NO is developed. A geometric–load graph based on the finite element mesh connectivity, where each node encodes spatial coordinates with local load conditions is constructed. A graph neural network is then employed to extract a unified global representation that captures both geometric structure and external force distribution. A query–decoder module combines global features with arbitrary query-coordinates to predict the full deformation
field. This query-based formulation enables accurate continuous deformation prediction across varying robot geometries, postures, and load conditions. Experimental results demonstrate that the proposed model achieves a root mean square error of 0.91 μm in deformation prediction and is 6000 times faster than conventional finite element simulations. This significant improvement enables rapid design, analysis and optimization for robotic mechanisms.