James Millar Ritchie
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
James Millar Ritchie has a total of 36 co-authors. Here are the top co-authors by number of shared publications:
- Theodore Lim
Heriot-Watt University
9 shared publications - Hugo Ivan Medellín Castillo
Facultad de Ingeniería, Universidad Autónoma de San Luis Potosí, México
5 shared publications - Xianwen Kong
School of Engineering and Physical Sciences, Heriot-Watt University, Edinburgh, U.K.
3 shared publications Andrew Brock
School of Engineering and Physical Sciences Heriot-Watt University Edinburgh UK
2 shared publications- Eder H. Govea-Valladares
Facultad de Ingeniería, Universidad Autónoma de San Luis Potosí, San Luis Potosí, Mexico
2 shared publications - J. B. C. Davies
Scottish Manufacturing Institute, Heriot-Watt University, Department of Mechanical Engineering, Riccarton, Edinburgh, Scotland, EH14 4AS
2 shared publications
Latest Publications
- 2 April 2020
- Published by Springer Nature in Virtual Reality
- Vol. 25 (1) , 53-67
Haptic-enabled virtual training in orthognathic surgery
- 2 April 2020
- Published by Springer Nature in Virtual Reality
- Vol. 25 (1) , 53-67
Haptic-enabled virtual training in orthognathic surgery
- Open Access
- 12 June 2018
- Published by Wiley in Journal of Computing and Information Science in Engineering
- Vol. 18 (3)
Automated Coordinate Measuring Machine Inspection Planning Knowledge Capture and Formalization
Capturing the strategy followed during a coordinate measuring machine (CMM) inspection planning session has been an extremely challenging issue due to the time-consuming nature of traditional methods, such as interviewing experts and technical documents data mining. This paper presents a methodology demonstrating how a motion capture-based system can facilitate direct and nonintrusive CMM operator logging for capturing planning strategies and representing in knowledge formats. With the use of recorded motion data, embedded knowledge and expertise can be captured automatically and formalized in various formats such as motion trajectory graphs, inspection plans, integrated definition (IDEF) model diagrams, and other representations. Additionally, a part program can be generated for driving a CMM to execute component measurement. The system's outputs can be used to help understand how a CMM inspection strategy is planned, as well as training aids for inexperienced operators and the rapid generation of part programs.
- 12 June 2018
- Published by Wiley in Journal of Computing and Information Science in Engineering
- Vol. 18 (3)
Open AccessAutomated Coordinate Measuring Machine Inspection Planning Knowledge Capture and Formalization
Capturing the strategy followed during a coordinate measuring machine (CMM) inspection planning session has been an extremely challenging issue due to the time-consuming nature of traditional methods, such as interviewing experts and technical documents data mining. This paper presents a methodology demonstrating how a motion capture-based system can facilitate direct and nonintrusive CMM operator logging for capturing planning strategies and representing in knowledge formats. With the use of recorded motion data, embedded knowledge and expertise can be captured automatically and formalized in various formats such as motion trajectory graphs, inspection plans, integrated definition (IDEF) model diagrams, and other representations. Additionally, a part program can be generated for driving a CMM to execute component measurement. The system's outputs can be used to help understand how a CMM inspection strategy is planned, as well as training aids for inexperienced operators and the rapid generation of part programs.
- Open Access
SMASH: One-Shot Model Architecture Search through HyperNetworks
Designing architectures for deep neural networks requires expert knowledge and substantial computation time. We propose a technique to accelerate architecture selection by learning an auxiliary HyperNet that generates the weights of a main model conditioned on that model's architecture. By comparing the relative validation performance of networks with HyperNet-generated weights, we can effectively search over a wide range of architectures at the cost of a single training run. To facilitate this search, we develop a flexible mechanism based on memory read-writes that allows us to define a wide range of network connectivity patterns, with ResNet, DenseNet, and FractalNet blocks as special cases. We validate our method (SMASH) on CIFAR-10 and CIFAR-100, STL-10, ModelNet10, and Imagenet32x32, achieving competitive performance with similarly-sized hand-designed networks. Our code is available at https://github.com/ajbrock/SMASH
Open AccessSMASH: One-Shot Model Architecture Search through HyperNetworks
Designing architectures for deep neural networks requires expert knowledge and substantial computation time. We propose a technique to accelerate architecture selection by learning an auxiliary HyperNet that generates the weights of a main model conditioned on that model's architecture. By comparing the relative validation performance of networks with HyperNet-generated weights, we can effectively search over a wide range of architectures at the cost of a single training run. To facilitate this search, we develop a flexible mechanism based on memory read-writes that allows us to define a wide range of network connectivity patterns, with ResNet, DenseNet, and FractalNet blocks as special cases. We validate our method (SMASH) on CIFAR-10 and CIFAR-100, STL-10, ModelNet10, and Imagenet32x32, achieving competitive performance with similarly-sized hand-designed networks. Our code is available at https://github.com/ajbrock/SMASH
- Open Access
- 24 March 2017
- Published by ASME International in Journal of Mechanisms and Robotics
- Vol. 9 (3)
Kinematic Analysis and Dimensional Synthesis of a Meso-Gripper
In recent years, applications in industrial assemblies within a size range from 0.5 mm to 100 mm are increasing due to the large demands for new products, especially those associated with digital multimedia. Research on grippers or robotic hands within the mesoscopic scale of this range has not been explored in any great detail. This paper outlines the development of a gripper to bridge the gap between microgrippers and macrogrippers by extending the gripping range to the mesoscopic scale, particularly without the need to switch grippers during industrial assembly. The mesoscopic scale gripper (meso-gripper) researched in this work has two modes of operation: passive adjusting mode and angled gripping mode, adapting its configuration to the shape of object automatically. This form of gripping and the associated mechanism are both novel in their implementation and operation. First, the concept of mesoscopic scale in robotic gripping is presented and contextualized around the background of inefficient hand switching processes and applications for assembly. The passive adjusting and angled gripping modes are then analyzed and a dual functional mechanism design proposed. A geometric constraint method is then demonstrated which facilitates task-based dimensional synthesis after which the kinematics of synthesized mechanism is investigated. The modified synthesized mechanism gripper is then investigated according to stiffness and layout. Finally, a 3D printed prototype is successfully tested, and the two integrated gripping modes for universal gripping verified.
- 24 March 2017
- Published by ASME International in Journal of Mechanisms and Robotics
- Vol. 9 (3)
Open AccessKinematic Analysis and Dimensional Synthesis of a Meso-Gripper
In recent years, applications in industrial assemblies within a size range from 0.5 mm to 100 mm are increasing due to the large demands for new products, especially those associated with digital multimedia. Research on grippers or robotic hands within the mesoscopic scale of this range has not been explored in any great detail. This paper outlines the development of a gripper to bridge the gap between microgrippers and macrogrippers by extending the gripping range to the mesoscopic scale, particularly without the need to switch grippers during industrial assembly. The mesoscopic scale gripper (meso-gripper) researched in this work has two modes of operation: passive adjusting mode and angled gripping mode, adapting its configuration to the shape of object automatically. This form of gripping and the associated mechanism are both novel in their implementation and operation. First, the concept of mesoscopic scale in robotic gripping is presented and contextualized around the background of inefficient hand switching processes and applications for assembly. The passive adjusting and angled gripping modes are then analyzed and a dual functional mechanism design proposed. A geometric constraint method is then demonstrated which facilitates task-based dimensional synthesis after which the kinematics of synthesized mechanism is investigated. The modified synthesized mechanism gripper is then investigated according to stiffness and layout. Finally, a 3D printed prototype is successfully tested, and the two integrated gripping modes for universal gripping verified.
- Open Access
Generative and Discriminative Voxel Modeling with Convolutional Neural Networks
When working with three-dimensional data, choice of representation is key. We explore voxel-based models, and present evidence for the viability of voxellated representations in applications including shape modeling and object classification. Our key contributions are methods for training voxel-based variational autoencoders, a user interface for exploring the latent space learned by the autoencoder, and a deep convolutional neural network architecture for object classification. We address challenges unique to voxel-based representations, and empirically evaluate our models on the ModelNet benchmark, where we demonstrate a 51.5% relative improvement in the state of the art for object classification.
Open AccessGenerative and Discriminative Voxel Modeling with Convolutional Neural Networks
When working with three-dimensional data, choice of representation is key. We explore voxel-based models, and present evidence for the viability of voxellated representations in applications including shape modeling and object classification. Our key contributions are methods for training voxel-based variational autoencoders, a user interface for exploring the latent space learned by the autoencoder, and a deep convolutional neural network architecture for object classification. We address challenges unique to voxel-based representations, and empirically evaluate our models on the ModelNet benchmark, where we demonstrate a 51.5% relative improvement in the state of the art for object classification.
- 5 July 2016
- Published by Springer Nature in The International Journal of Advanced Manufacturing Technology
- Vol. 89 (1-4) , 581-598
The analysis and evaluation of the influence of haptic-enabled virtual assembly training on real assembly performance
- 5 July 2016
- Published by Springer Nature in The International Journal of Advanced Manufacturing Technology
- Vol. 89 (1-4) , 581-598
The analysis and evaluation of the influence of haptic-enabled virtual assembly training on real assembly performance
- 1 July 2016
- Published by Elsevier in Computer Methods and Programs in Biomedicine
- Vol. 130, 46-53
The evaluation of a novel haptic-enabled virtual reality approach for computer-aided cephalometry
- 1 July 2016
- Published by Elsevier in Computer Methods and Programs in Biomedicine
- Vol. 130, 46-53
The evaluation of a novel haptic-enabled virtual reality approach for computer-aided cephalometry
- 24 November 2015
- Published by ASME International in Journal of Mechanisms and Robotics
- Vol. 8 (2) , 024501
Determination of the Workspace of a Three-Degrees-of-Freedom Parallel Manipulator Using a Three-Dimensional Computer-Aided-Design Software Package and the Concept of Virtual Chains1
The determination of workspace is an essential step in the development of parallel manipulators. By extending the virtual-chain (VC) approach to the type synthesis of parallel manipulators, this technical brief proposes a VC approach to the workspace analysis of parallel manipulators. This method is first outlined before being illustrated by the production of a three-dimensional (3D) computer-aided-design (CAD) model of a 3-RPS parallel manipulator and evaluating it for the workspace of the manipulator. Here, R, P and S denote revolute, prismatic and spherical joints respectively. The VC represents the motion capability of moving platform of a manipulator and is shown to be very useful in the production of a graphical representation of the workspace. Using this approach, the link interferences and certain transmission indices can be easily taken into consideration in determining the workspace of a parallel manipulator.
- 24 November 2015
- Published by ASME International in Journal of Mechanisms and Robotics
- Vol. 8 (2) , 024501
Determination of the Workspace of a Three-Degrees-of-Freedom Parallel Manipulator Using a Three-Dimensional Computer-Aided-Design Software Package and the Concept of Virtual Chains1
The determination of workspace is an essential step in the development of parallel manipulators. By extending the virtual-chain (VC) approach to the type synthesis of parallel manipulators, this technical brief proposes a VC approach to the workspace analysis of parallel manipulators. This method is first outlined before being illustrated by the production of a three-dimensional (3D) computer-aided-design (CAD) model of a 3-RPS parallel manipulator and evaluating it for the workspace of the manipulator. Here, R, P and S denote revolute, prismatic and spherical joints respectively. The VC represents the motion capability of moving platform of a manipulator and is shown to be very useful in the production of a graphical representation of the workspace. Using this approach, the link interferences and certain transmission indices can be easily taken into consideration in determining the workspace of a parallel manipulator.
- 2 September 2015
- Published by Springer Nature
- p. 16-30
To Facilitate or Not? Understanding the Role of the Teacher in Using a Serious Game
- 2 September 2015
- Published by Springer Nature
- p. 16-30
To Facilitate or Not? Understanding the Role of the Teacher in Using a Serious Game
- 1 March 2011
- Published by Springer Nature in Virtual Reality
- Vol. 16 (1) , 1-2
Editorial