Technology Knowledge Graph for Design Exploration: Application to Designing the Future of Flying Cars

To pursue innovation, design engineers need to continuously exploit the knowledge in their design domain and explore other relevant knowledge around the domain. While many methods and tools have been developed to retrieve knowledge within a given design domain, e.g., flying cars, knowledge discovery beyond the domain for innovation remains a challenge, and relevant methods are under-developed. Herein, we introduce a methodology to use a technology knowledge graph (TKG), which covers sematic-level knowledge in all technology fields defined in the international patent classification system, to retrieve the existing engineering knowledge in a domain and discover engineering concepts around the domain for future design considerations and innovation. We demonstrate the TKG-based methodology by applying it to explore the future designs of flying cars, an emerging domain with high uncertainty despite growing prospects.

Towards a Broad Use of Gamification Based on Hand Tracking in Post Stroke Patients

In the last years, the advent of innovative technologies for tracking human motions is increasing the interest of physicians and physiotherapist, who would like to introduce new instruments for a more objective assessment of the rehabilitation processes. At present, many motion tracking systems have been developed and their ease of use and low-cost may represent the key aspects for which these systems could be really adopted both in rehabilitation centers and in rehabilitation programs at home. Several research studies confirmed the importance of continuing rehabilitation programs at home with the aim to maintain patients’ health condition at a suitable level for daily life activities. Physicians and physiotherapists need methods and tools, which can be simply adaptable for each type of patients’ category and type of rehabilitation according to the assessed pathology. For achieving this need, the technology has to be suitable for both the patient side and medical personnel side. The most suitable technology for the patients are motion tracking devices which can be used through traditional IT, such as laptops, smartphones and tablets. Also for medical personnel the ease of use is very important, physicians would like to check the patient’s rehab exercises according to their medical knowledge by exploiting daily life technology. This research work investigates on which are the best user-friendly programming tools and low-cost technology for 3D hand and finger tracking for the development of a serious game for rehabilitation exercises. The tasks are designed according to physiotherapists’ recommendations, in order to be customizable for any single user. The following sections will describe the method, the tools adopted, and the application developed.

Towards Selective Volumetric Additive Manufacturing and Processing of Ceramics

The development of advanced additive manufacturing (AM) and material processing techniques is currently a topic of great interest to broad communities of scientists and engineers. In particular, there is a need for AM processes capable of producing functional and high-quality components at a faster rate than is currently achievable. In response to this demand, the present work introduces the initial steps of a novel spatially-resolved and selective approach for processing volumetric regions of ceramic materials. The proposed method utilizes microwave radiation to heat material at desired locations within a domain filled with ceramic powder. Using this principle of operation, a number of methods for implementation of this process are proposed. As a first step, a multiphysics computational methodology and an associated model that allows for the analysis and design of relevant processing systems is introduced. Additionally, a number of simulations demonstrating the feasibility of the proposed methodology are presented. Based on these preliminary results, we conclude with a discussion of ongoing and future efforts to fully realize this technology.

Do Autonomous Vehicle Driving Styles Affect User State?: A Preliminary Investigation

A bursting number of achievements in the autonomous vehicle industry have been obtained during the past decades. Various systems have been developed to make automated driving possible. Due to the algorithm used in the autonomous vehicle system, the performance of the vehicle differs from one to another. However, very few studies have given insight into the influence caused by implementing different algorithms from a human factors point of view. Two systems based on two algorithms with different characteristics are utilized to generate the two driving styles of the autonomous vehicle, which are implemented into a driving simulator in order to create the autonomous driving experience. User’s skin conductance (SC) data, which enables the evaluation of user’s cognitive workload and mental stress were recorded and analyzed. Subjective measures were applied by filling out Swedish occupational fatigue inventory (SOFI-20) to get a user self-reporting perspective view of their behavior changes along with the experiments. The results showed that human’s states were affected by the driving styles of different autonomous systems, especially in the period of speed variation. By analyzing users’ self-assessment data, a correlation was observed between the user “Sleepiness” and the driving style of the autonomous vehicle. These results would be meaningful for the future development of the autonomous vehicle systems, in terms of balancing the performance of the vehicle and user’s experience.

Towards Thermal Simulation of Powder Bed Fusion on Path Level

Powder bed fusion (PBF) is a widely used additive manufacturing (AM) technology to produce metallic parts. Understanding the relationships between process parameter settings and the quality of finished parts remains a critical research question. Developing this understating involves an intermediate step: Process parameters, such as laser power and scan speed, influence the ongoing process characteristics, which then affect the final quality of the finished parts. Conventional approaches to addressing those challenges such as powder-based simulations (e.g., discrete element method (DEM)) and voxel-based simulations (e.g., finite element method (FEM)) can provide valuable insight into process physics. Those types of simulations, however, are not well-suited to handle realistic manufacturing plans due to their high computational complexity. Thermal simulations of the PBF process have the potential to implement that intermediate step. Developing accurate thermal simulations, however, is difficult due to the physical and geometric complexities of the manufacturing process. We propose a new, meso-scale, thermal-simulation, which is built on the path-level interactions described by a typical process plan. Since our model is rooted in manufactured geometry, it has the ability to produce scalable, thermal simulations for evaluating realistic process plans. The proof-of-concept simulation result is validated against experimental results in the literature and experimental results from National Institute of Standards and Technology (NIST). In our model, the laser-scan path is discretized into elements, and each element represents the newly melted material. An element-growth mechanism is introduced to simulate the evolution of the melt pool and its thermal characteristics during the manufacturing process. The proposed simulation reduces computational demands by attempting to capture the most important thermal effects developed during the manufacturing process. Those effects include laser-energy absorption, thermal interaction between adjacent elements and elements within the underneath substrate, thermal convection and radiation, and powder melting.

Construction of Multi-Domain System Simulation Model for Trade Studies in System Design Considering Multiple Scenes

As a product is sold globally, usages of the product have much wider variety. Thus, a product needs to be designed considering multiple scenes. To certify that the product performs properly in any scene, industries started to apply Model Based Systems Engineering (MBSE). Whereas multi-domain system simulations are regarded as a prominent approach for the system design of a product, construction of model depends on knowledge and sense modelers. This paper proposes a modelling method to construct appropriate multi-domain system simulation models while reducing dependencies to senses of modelers. The proposed method comprises two parts. First, significant tradeoffs to be studied by the simulation are specified. Second, features of simulation models are deliberated for specified tradeoffs. To specify significant tradeoffs, product and scenes where the product is used are integrated into a model. Further, to deliberate features of simulation model, cognitive model of physical phenomena in a product is employed as well. The proposed method was applied to the development of continuously variable transmission to verify its validity.

Mahalanobis-Taguchi-Fukuda Approach to Human Motion Control

Learning from failures approach how to control human motion is developed by extending Mahalanobis Taguchi System. It enables quantitative measurement of how the learner is improving in his or her learning, It helps to acquire tacit knowledge such as swimming, for which we do not have valid approach. Since Mahalanobis Distance is a unitless measure for multi-dimensional variables, this approach can be extended to many adaptive network formation and management, because this approach let the learner recognize how he or she can coordinate their body pars to adapt to the changing situations. Thus, the approach ca be applied to development and operation of soft robots and adaptive network or team formation and management in the IoT connected society.

Dimensional Prediction for FDM Machines Using Artificial Neural Network and Support Vector Regression

With the development of Fused Deposition Modeling (FDM) technology, the quality of fabricated parts is getting more attention. The present study highlights the predictive model for dimensional accuracy in the FDM process. Three process parameters, namely extruder temperature, layer thickness, and infill density, are considered in the model. To achieve better prediction accuracy, three models are studied, namely multivariate linear regression, Artificial Neural Network (ANN), and Support Vector Regression (SVR). The models are used to characterize the complex relationship between the input variables and dimensions of fabricated parts. Based on the experimental data set, it is found that the ANN model performs better than the multivariate linear regression and SVR models. The ANN model is able to study more quality characteristics of fabricated parts with more process parameters of FDM.

Pixel to Stroke Sketch Generation Using Reinforcement Learning

Many engineering design tasks involve creating early conceptual sketches that do not require exact dimensions. Although some previous works focus on automatically generating sketches from reference images, many of them output exactly the same objects as the reference images. There are also models that generate sketches from scratch, which can be divided into pixel-based and stroke-based methods. Pixel-based methods generate sketches as a whole, without any information of the strokes, while stroke-based methods generate sketches by outputting strokes in a sequential manner. Pixel-based methods are frequently used to generate realistic color images. Although the pixel-based methods are more popular, stroke-based methods have the advantages to scale to a larger dimension without losing high fidelity. An image generated from stroke-based methods has only strokes on the canvas, resulting in no random noise in the blank areas of the canvas. However, one challenge in the engineering design community is that most of the sketches are saved as pixel-based images. Furthermore, many non-pixel-based methods rely on stroke-based training data, making them ill-suited for generating design conceptual sketches. In order to overcome these limitations, the authors proposed an agent that can learn from pixel-based images and generate stroke-based images. An advantage of such an agent is the ability to utilize pixel-based training data that is abundant in design repositories, to train stroke-based methods that are typically constrained by the lack of access to stroke-based training data.

Using Virtual Reality to Test Human-Robot Interaction During a Collaborative Task

The use of collaborative robots in the manufacturing industry has widely spread in the last decade. In order to be efficient, the human-robot collaboration needs to be properly designed by also taking into account the operator’s psychophysiological reactions. Virtual Reality can be used as a tool to simulate human-robot collaboration in a safe and cheap way. Here, we present a virtual collaborative platform in which the human operator and a simulated robot coordinate their actions to accomplish a simple assembly task. In this study, the robot moved slowly or more quickly in order to assess the effect of its velocity on the human’s responses. Ten participants tested this application by using an Oculus Rift head-mounted display; ARTracking cameras and a Kinect system were used to track the operator’s right arm movements and hand gestures respectively. Performance, user experience, and physiological responses were recorded. The results showed that while humans’ performances and evaluations varied as a function of the robot’s velocity, no differences were found in the physiological responses. Taken together, these data highlight the relevance of the kinematic aspects of robot’s motion within a human-robot collaboration and provide valuable insights to further develop our virtual human-machine interactive platform.