Joseph Jayakody

GRADUATING IN FALL 2026, I'M ACTIVELY LOOKING FOR NEW POSITIONS.

Research Interests: Digital manufacturing, Shape optimization, AI-driven design optimization, Machine learning, Geometric computing

I am a PhD candidate in Computational Design at the EPSRC CDT in Topological Design, within the School of Physics and Astronomy of the University of Birmingham, advised by Hyunyoung Kim, David Lee Butler and Lauren E.J. Thomas-Seale. During my PhD, I completed an academic placement under the supervision of Bailin Deng at Cardiff University. My current work is at the intersection of geometric modelling, shape optimization and AI-driven design, with a focus on addressing shortcomings inherent to additive manufacturing processes.

In 2022, I graduated from The Hong Kong University of Science and Technology (HKUST) with a MPhil in Smart Manufacturing, where I worked on curved-layer process planning for multi-axis additive manufacturing, under the supervision of Ravindra Goonetilleke and Kai Tang.

Even prior to that, I graduated from Sheffield Hallam University with a BEng (Hons) degree in Electronic Engineering and was honoured to receive the Pro Vice Chancellor's award for the Best Performer - Class of 2019.

I'm open to new collaborations and opportunities. Please reach out to me through dxj219@student.bham.ac.uk.

Publications

Neural field-based shape optimization for manufacturability-aware support structure minimisation

Authors: Don Pubudu Vishwana Joseph Jayakody, Bailin Deng, Ravindra Goonetilleke , Lauren Thomas-Seale and Hyunyoung Kim

Journal: Additive Manufacturing

Paper link , Project page

Abstract: Support structure requirement for overhang regions is perhaps the most critical constraint in material extrusion-based additive manufacturing (AM), which increases material waste and print time, leading to high production cost. Direct shape optimisation eliminates the need to manually refine or redesign complex geometries to meet manufacturability requirements, including minimising support structures. However, direct mesh-based shape manipulation (e.g., rigid rotations) methods are not only limited in terms of part complexity and mesh resolution, but also inherent to large deviation and surface feature distortions.

In this paper, we present a novel AM-oriented, end-to-end neural shape optimisation framework to minimise (if not eliminate) overhang regions of a range of complex geometric models with non-trivial topology and intricate surface features. Our method learns the optimal geometric deformation using a neural field which is governed by a set of manufacturability-oriented loss functions. This mesh-free approach realises overhang minimisation under negligible surface feature distortion, and minimal deviation for all the tested geometries. By leveraging the smoothness and continuity of the neural field, we then introduce a coarse-to-fine optimisation workflow to realise direct and efficient optimisation of high resolution meshes. The proposed approach is validated through extensive computational and physical printing experiments. Our results show an average reduction of 46% in support-structure print time, including a case in which support-structure print time is reduced by 100% for a complex geometry, which clearly demonstrate the effectiveness of the proposed computational framework and its potential as a strong foundation towards AI-driven design to achieve support-free extrusion-based additive manufacturing.

Keywords: AI-driven design, Geometric optimization, Support-free 3D printing, Design for additive manufacturing (DfAM), AI-driven 3D printing, Neural shape editing

Part orientation fused shape optimization for minimisation of print time and material waste in extrusion-based 3D printing

Authors: Don Pubudu Vishwana Joseph Jayakody, Bailin Deng, Ravindra Goonetilleke , Lauren Thomas-Seale and Hyunyoung Kim

Journal: Computer-Aided Design

Paper link , Project page

Abstract: Support structure generation is a critical requirement in additive manufacturing (AM) to prevent material collapse in overhanging regions. However, it increases print time, material waste, and overall production cost, especially in extrusion-based AM. To mitigate these problems, design engineers often resort to manually finetuning or even redesigning prototype geometry to minimise support structures, which is time-consuming and inefficient. A direct geometric optimisation that preserves locality of shape changes whilst corresponding to the part orientation remains an underdetermined problem.

In this paper, we present a novel alternating optimisation framework that finds the corresponding part geometry and orientation to minimise support structures under minimal geometric deviation. Whilst global-level support structure reduction is realised by the part orientation change, we introduce an efficient energy minimisation-based geometric optimisation framework, which is governed by saliency-aware elementwise projections and a set of manufacturing constraints. The proposed framework is validated through extensive computational and physical printing experiments employing multiple 3D printers and support structure types, on a diverse set of complex models including topologically non-trivial parts such as gyroid structures. Our results show an average reduction of 50 % in support structure print time, 27 % in material usage and 25 % in total print time, demonstrating the effectiveness of the proposed framework and its potential as a paradigm shift in manufacturing-oriented design.

Keywords: Shape optimization, Part orientation optimization, Design optimization, DfAM, Design for Additive Manufacturing

A salient vector field-driven part orientation selection for multi-axis 3D printing

Authors: Don Pubudu Vishwana Joseph Jayakody, Tak Yu Lau , Hyunyoung Kim, Kai Tang and Lauren Thomas-Seale

Journal: Computer-Aided Design

Paper link , Project page

Abstract: Part orientation is a crucial element that governs the impact of several manufacturing constraints in material extrusion-based additive manufacturing (AM). Although part orientation optimisation has been extensively investigated to improve the manufacturability in 2.5-axis AM configuration, its influence on material extrusion-based multi-axis AM remains underdetermined. In this paper, we propose a computational framework to find the optimal part orientation that maximises the compliance of the tool orientation vector field with respect to several constraints required for support-free multi-axis AM. By combining topological significance, mesh saliency and curvedness metrics, we introduce a new salient feature map to formulate the link between the part orientation and the tool orientation vector field compliance. Once the optimal orientation is computed, our method enables a direct computation of a compliant iso-tool orientation vector field for a set of input iso-tool path points. We demonstrate that the part orientation can indeed be changed to minimise tool angle variation whilst adhering to overhang angle constraints for a range of 3D mesh models. The effectiveness of the proposed method is validated by comparing our method with existing tool orientation vector field design methods. Our promising results reveal the potential in part orientation optimisation as a means to address manufacturing constraints in multi-axis tool path design.

Keywords: Part orientation optimisation, Saliency, Mesh saliency, Reeb graph, Morse theory, Critical points, Saliency map, Multi-axis AM, Multi-axis 3D printing, Curved slicing, Pattern search, Vector field design, Vector field optimisation

Topological awareness towards collision-free multi-axis curved layer additive manufacturing

Authors: Don Pubudu Vishwana Joseph Jayakody, Tak Yu Lau , Hyunyoung Kim, Kai Tang and Lauren Thomas-Seale

Journal: Additive Manufacturing

Paper link

Abstract: We present a topology-aware computational framework that enables global collision-free and support-structure-free material extrusion for curved layer multi-axis printing of a range of high-genus and simply connected models. The nature of multi-axis additive manufacturing facilitates continuous variation of build direction, which paves the way for numerous approaches of geometry-based curved layer design. However, due to unawareness of global topological features, prevention of collision between the printer nozzle and the workpiece can either remain uncertain or demand for exhaustive algorithmic approaches, depending on the topological and geometrical complexity of a given mesh model. In this paper, we present a topological analysis framework that draws inspiration from the concepts of Reeb graph and Morse theory, to distinguish between collision-prone and collision-free regions of mesh models. We provide an exact definition for global collision-inducing features in regions of risk and optimise the nozzle orientation vector field to avoid global collisions whilst adhering to the overhang angle constraints. A new shape analysis method is then proposed to find the model orientation which maximises the manufacturability of bifurcating simply connected models. To validate our algorithmic framework, several high-genus and bifurcating simply connected models are printed using a robotic multi-axis printer. The experimental results demonstrate the feasibility and effectiveness of our framework for minimising global collisions in multi-axis curved layer additive manufacturing for a range of mesh models.

Keywords: Geometric computing, Computer aided design, Computational design, 3D printing, Multi-axis printing, Tool-axis vectors, Tool orientation, Nozzle vector, Topology, Computational topology, Process planning, Collision-free, Surface quality, Reeb graph, Morse theory, Critical points.

Convexity and surface quality enhanced curved slicing for support-free multi-axis fabrication

Authors: Don Pubudu Vishwana Joseph Jayakody, Tak Yu Lau ,Ravindra Stephen Goonetilleke and Kai Tang

Journal: Journal of Manufacturing and Materials Processing

Paper link, Project page

Abstract: In multi-axis fused deposition modeling (FDM) printing systems, support-free curved layer fabrication is realized by continuous transition of the printer nozzle orientation. However, the ability to print 3D models with complex geometric (e.g., high overhang) and topological (e.g., high genus) features is often restricted by various manufacturability constraints inherent to a curved layer design process. The crux in a multi-axis printing process planning pipeline is to design feasible curved layers and their tool paths that will satisfy both the support-free condition and other manufacturability constraints (e.g., collision-free). In this paper, we propose a volumetric curved layer decomposition method that strives to not only minimize (if not prevent) collision-inducing local shape features of layers, but also enable adaptive layer thickness to comply with a new volumetric error-based surface quality criterion. Our method computes an optimal Radial Basis Functions (RBF) field to modify the fabrication sequence field, from which, the iso-surface layers are extracted to design the corresponding multi-axis printing tool paths. A method to fine-tune variable nozzle orientations on each resulting curved layer is then proposed to overcome possible collisions in high-genus geometries. To validate the concept and exhibit its potential, several support-free fabrication experiments and comparisons with the conventional geodesic field-based slicing are presented, and the results give a preliminary confirmation of the feasibility and advantages of the proposed method.

Keywords: Robotic manufacturing, Geometry, Convexity, Concavity, Geometric computing, Computer aided design, Curvature, Principal curvatures, Discrete differential geometry, Support-structures, Collision

Curved layer process planning for robotic multi-axis additive manufacturing

Author: Don Pubudu Vishwana Joseph Jayakody

MPhil Thesis - Hong Kong University of Science and Technology

Thesis link

Abstract: Additive Manufacturing (AM) technology offers tremendous flexibility in fabricating complex geometries and low material wastage in contrast to conventional manufacturing techniques. However, the layer-by-layer build up process and the fixed build direction native to the AM technology cause pronounced staircase effect and requirement of support-structures in fabricating 3D geometries with large overhang areas, thus limiting the true potential of AM. The initial concept of curved multi-axis AM emerged to realise support-structure-free fabrication by exploiting the enhanced build direction flexibility. Since this method allows fabricating along non-planar (i.e., curved) tool paths with a continuously changing nozzle orientation, the possibility of support-free fabrication of geometries with minimised staircase effect has been evident in theory. However, in practice, multi-axis fabrication suffers from a number of manufacturing constraints that diminish the complexity of geometries that could be fabricated in a support-free manner. Local gouging (i.e., collisions induced between the printer nozzle and the printed portions) has been an inevitable constraint that is native to the concave features of the decomposed surface layers. Efforts that have been taken to improve the convexity (i.e., eradicate concavity) in decomposed surface layers have either failed in maintaining support-free quality in curved fabrication or suffered with severe surface artifacts and approximation errors. Furthermore, there has been minimal effort towards enhancing the fabrication surface quality (i.e., surface finish) in 3D geometries fabricated through curve layer decomposition. This thesis presents a new volumetric curved layer decomposition method that generates concavity reduced (if not eradicated) and surface quality improved curved surface layers while maintaining overhang angles below the support-structure-free thresholds to realise multi-axis fabrication. Mathematical modelling frameworks for surface concavity identification, volumetric error approximation and overhang angle condition monitoring are developed to evaluate the impact of manufacturing constraints within a set of decomposed layers. Based on the severity of these impacts, a weighted gradient vector field that represents the 3D volume of the given model is designed by employing the Radial Basis Functions (RBF). Weights are optimised through a genetic algorithm-based optimisation model to find the ideal RBF parameters that would generate the best possible set of decomposed surface layers minimising the manufacturing constraints. Finally, a contour parallel multi-axis fabrication tool path is designed on each decomposed curved surface layer to realise physical fabrication. A novel strategy to plan the nozzle orientation vector is presented to minimise collisions and to achieve safe fabrication. To validate the presented curve layer decomposition method and the tool path, simulations and physical experiments are performed with selected 3D geometries that contain not only complex geometrical features (i.e., high-genus) but also high overhang areas.

Keywords: Robotics, Curved layer slicing, Geometric computing, Computer aided design, Computer aided manufacturing, Computational design, 3D printing, Multi-axis printing

Page updated

Google Sites

Report abuse