Moohyun Cha

An interactive data-driven driving simulator using motion blending

Compared to the motion equations the data-driven method can simulate reality from sampling of real motions but real-time interaction between a user and the simulator is problematic. Existing data-driven motion generation methods simply record and replay the motion of the vehicle. Character animation technology enables a user to control motions that are generated by a motion capture database and an appropriate motion control algorithm. We propose a data-driven motion generation method and implement a driving simulator by adapting the method of motion capture. The motion data sampled from a real vehicle are transformed into appropriate data structures called motion blocks, and then a series of motion blocks are saved into the motion database. During simulation, the driving simulator searches for and synthesizes optimal motion blocks from the motion database and generates motion streams that reflect the current simulation conditions and parameterized user demands. We demonstrate the proposed method through experiments with the driving simulator.

A framework for a multi-sensory VR effect system with motional display

Virtual reality simulators have been developed as tools for the transfer of knowledge and education. Concurrently, the demand for exhibition systems of science education and cultural experiences has also increased. Existing VR (virtual reality) simulators, which are based on the calculation of dynamics equations, cannot easily be adapted to changes in simulation content. In order to transfer knowledge and maintain interest through educational applications, an experiential system that has multi-sensory effects as well as motion effects is needed. The authors proposed a method for motion generation that is tailored to riding systems and multi-sensory VR effects. In this study, both sense of motion, which is generated from movement of the viewpoint of the visual image, and motion effects, which are prepared in advance, are blended to realize motion simulation by the proposed method. Motion effects can easily be added by interaction between the user and the riding system. Various sensory cues are also implemented to the riding system

A Hybrid Driving Simulator with Dynamics-Driven Motion and Data-Driven Motion

In driving simulator technology, which reproduces the sense of motion in a virtual reality environment, various research activities are attempting to enhance realism. Recently studies have focused on a data-driven method of generating motion; this method records the motions of real objects and regenerates them in a simulation system. By providing a simulator with real motion data, this method can easily guarantee realism without complicated dynamics models and equation solvers. This paper proposes a hybrid method of generating controllable motion in a driving simulator through the application of a data-driven motion generation method and a dynamics-based motion generation method. The driving data acquired from a real vehicle are parsed into motion blocks with a dynamic model of the vehicle; the motion blocks are then stored in the database. In real-time simulation, a natural motion stream is generated by a process of searching for and synthesizing optimal motion blocks from the database and is synthesized with dynamics-based motion again. A more realistic sense of motion can be generated by parameterizing the current simulation state and user requirements.

A Walking Movement System for Virtual Reality Navigation

A walking navigation system (usually known as a locomotion interface) is an interactive platform which gives simulated walking sensation to users using sensed bipedal motion signals. This enables us to perform navigation tasks using only bipedal movement. Especially, it is useful for the certain VR task which emphasizes on physical human movement, or accompanies understanding of the size and complexity of building structures. In this work, we described system components of VR walking system and investigated several types of walking platform by literature survey. We adopted a MS Kinect depth sensor for the motion recognition and a treadmill which includes directional turning mechanism for the walking platform. Through the integration of these components with a VR navigation scenario, we developed a simple VR walking navigation system. Finally several technical issues were found during development process, and further research directions were suggested for the system improvement.

A distributed visualization module and its applications using tiled display wall

In recent years, the use of large high-resolution display systems, usually implemented through tiled display walls, has been increasing in many areas of industry. In the virtual reality area, distributed visualization using PC clusters is generally used and it is necessary to design a synchronization procedure and to define appropriate simulation parameters to be synchronized within it. In this work, we developed a distributed visualization module to synchronize tiled scene contents generated by the open-source graphics engine, using the consistent shared-object mechanism which can be interoperable with the application-specific scene graph. We then applied this module to a couple of virtual reality applications using a DLP-Cube based tiled display system in order to evaluate its visualization result and real-time performance.

A data-driven visual simulation of fire phenomena

In order to simulate and visualize natural phenomena, especially fluid behavior such as smoke and fire, many novel studies have recently been conducted. Usually these methods use CFD (computational fluid dynamics), which calculate Navier-Stokes equations in real-time to generate realistic fluid motion and interactions, as well as high-performance GPU technologies. We proposed a new approach to the visual simulation of fluid flow by combining the use of pre-calculated CFD data with the real-time processing of such data. As the domain-specialized CFD solver predicts detailed fluid dynamics to an accuracy of a guaranteed error range, we could provide nearly actual behaviors of a fire-driven fluid flow. Moreover, this CFD data includes physical quantities such as temperature distribution, which can provide useful information to the training evaluation process. However, the data-driven method requires appropriate data processing techniques to create and manage large data sets. In this study, we developed a firefighter training simulator to demonstrate our proposed methods and explore related research issues.

A team-based firefighter training platform using the virtual environment

The increasing complexity of modern buildings, such as high-rise buildings and underground subway stations, has brought about an increase in the scale of potential hazards, which in turn requires changes in firefighter training. Due to cost, time, and safety requirements for the trainee, it is impossible for firefighters in training to experience a real fire. In additional, repeated training is impossible if real fire were used. We proposed a team-based firefighter training platform using the virtual environment for complex buildings. The platform provides training and evaluation of firefighting and mission-based team training. To achieve an immersive virtual environment, VR, AR and haptics technique is adopted. A numerical analysis method is used to model real fire phenomena.

Cluster rendering on large high-resolution multi-displays using X3DOM and HTML

In this paper, we propose a platform-independent visualization method that uses HTML and X3DOM for large high-resolution displays. HTML is a platform-independent language used for developing web pages, and X3DOM is a Web 3D visualization framework based on X3D; X3D is a royalty-free ISO standard XML-based file format for representing a 3D scene graph. Large high-resolution displays are used in scientific visualization, immersive virtual reality environment, and collaborative designing to efficiently deliver a large amounts of information. We introduce a run-time cluster configuration method using the component technologies of HTML5 and a data distribution strategy for a cluster rendering system based on the X3DOM framework. The comparative performance analysis and visualization results in diverse large high-resolution displays and mobile devices using widespread web browsers are also introduced. Our approach provides a cluster rendering method that can be easily and readily applied for existing large high-resolution systems, involving various platforms, by enabling novel system architecture.

Visualizing and experiencing harmful gases in the VR environment

We present an approach to visualizing and experiencing the harmful gases caused by fire in the virtual reality environment. It is known that a large portion of the casualties that result from fires in an enclosed space are due to poisoning by noxious gases rather than serious burns. The virtual-reality based visualization of those harmful gases will help people to enhance first emergency response and increase human safety. Because it is not currently possible to simulate the propagation of harmful gases in real time, we numerically simulate the behavior of fire-driven gases using an appropriate CFD solver off-line, then visualize the resulting voxel data through direct volume rendering, enabling an intuitive experience of gases in a virtual 3D space. Also, to visualize this type of naturally invisible gas data, we designed a transfer function based on the safety or danger range of gases, which can be customized by fire engineers or training supervisors.

Erratum to: Shape distribution-based retrieval of 3D CAD models at different levels of detail

With increasing design reuse in modern products, accurate and efficient 3D CAD model retrieval methods are required. To improve retrieval capability, a 3D shape comparison method is often utilized. In this method, the shapes of 3D CAD models in a database are compared with the shape given by a user. Meanwhile, for rapid generation of query models, a freehand sketch-based modeling method is adopted for retrieval systems. This method creates a low-level-of-detail (low-LOD) 3D CAD model with abbreviated exterior shapes. On the other hand, the target 3D CADmodel in the database is a high-LOD 3D CADmodel including detailed shapes of a product or components of a product. Considering different LODs of query and target models, we propose a new 3D CAD model retrieval method consisting of a 3D CAD model simplification system and a shape distribution-based shape comparison engine that compares multi-resolution models in a database to improve retrieval accuracy using a query model with simple shape. Experiment is conducted on 64 LOD models generated from 8 test cases and 8 query models generated by freehand sketch-based modelling method. Result shows a 200 % improvement on retrieval success rate for lower LOD models (100 %) compared with source models (50 %). Moreover, the proposed method has an advantage on efficiency, due to the simple calculation method and short computation time. Multimed Tools Appl DOI 10.1007/s11042-016-3881-5 * Duhwan Mun dhmun@knu.ac.k Hyungki Kim diskhkme@gmail.com Moohyun Cha mhcha@kimm.re.kr 1 Mechanical Systems Safety Research Division, Korea Institute of Machinery and Materials, 156, Gajeongbuk-ro, Yuseong-gu, Daejeon 305-701, South Korea 2 Department of Precision Mechanical Engineering, Kyungpook National University, 2559, Gyeongsang-daero, Sangju, Gyeongsangbuk-do 742-711, South Korea