Most of teleoperation applications take place in remote environments that are unstructured and not precisely known a priori. Therefore, fully autonomous control of an unmanned system is infeasible in such cases. To impose human intelligence on the task to cope with uncertainties, an improved telepresence system that eliminates the gap between machine and operator is essential. Current methods mainly use two-dimensional (2D) image covering a restricted field-of-view (FOV) as visual aid and numeric onboard sensor readings as references for operators. Since, sensory cues are significantly lost which include ambient visual information, depth information, and kinesthetic input. Also, in cases that use unmanned vehicle system (UVS) to inspect fully unknown areas, sensor information is not applicable until it is analyzed or processed. Instead of sending multiple sensors into the field with unmanned vehicles and retrieve a large amount of data for time-consuming post analysis, an instinct and sensory method for quick understanding and reconstruction of onsite situation is needed.

The major effort of this research is taking advantage of state-of-art technologies of three-dimensional (3D) input/output and developing an avatar-like mechanism to synchronize the physical behavior of an operator. Two of 3D input/output methods are used in this research: stereoscopic vision and motion tracking. The system is designed to work under a close-loop control with human feedback involved. Two cameras that are optimized according to human eyes angle-of-view (AOV) for accommodating human eyes’ perspective are used as stereoscopic vision input on the unmanned vehicle. The stereoscopic image is then streamed back to the operator through radio downlink in realtime. On the operator end, a head mount display (HMD) is used for displaying the stereoscopic image and tracking head movement of the operator with embedded sensors. The head tracking data are interpreted to control signal and returned to unmanned vehicle for controlling a three-axis gimbal mechanism on which the two cameras are installed. Then the loop is closed by synchronizing the operator and cameras’ FOV.

Medi connects elder patients and physical pharmacies closer by utilizing numerous home automation methods to their medication routines. Medi is a cloud medication management service with a physical home unit for each patient.

How It Works?

As the patient got a prescription from the doctor, the hospital sent it to the Medi Cloud, a virtual pharmacy ran by physical pharmacies. In the meanwhile, the patient went home and wait for the delivery of Medi Roll, the redesigned pre-sorted and easy-to-tear pill packets, would be delivered within a day. The patient refilled the Medi Home unit with the new arrived Medi Roll, set the time of his/hers four daily routine times: breakfast, lunch, dinner, sleep times, and Medi Home would took care from here. The patients would be notified of taking medication and given the exact pill packet automatically, whether the instruction of prescription requires patient to take right after lunch, every four hour, or before sleep. When the patient went outdoors, Medi Go, a synced portable keychain, would notify the patient like Medi Home did.

Medi Roll already had all prescribed medication schedule programmed inside and every packet is sorted according to the schedule. Medi Home “customized” the schedule through preprogrammed mechanism to fit the life schedule of the individual patient. Once the pill packets rans out, Medi Home connected to Medi Cloud to request for further medication instruction, whether asking for same prescription delivery or notifying patient for returning to the doctor.

Vision

Medi solved the two major problem of elder’s medication as we addressed in advance: complicated instructions and a lack of awareness on scheduling. Also, many minor problems were solved such as difficulties on sorting, preserving, carrying pills, and more medication tracking and data analysis.

Medi makes all the ways you’re used to taking medication more simple, convenient, and reliable. It is a whole new life experience to be released from annoying pilling routines. One that’s simpler than ever. We believe that care providers are desperately needed in these processes to confirm that patients are getting their medications on time and in the correct doses.

UAVAR is an AR system including hardware and software focusing on applying UAV in construction scenarios. Augmented reality (AR), the technologies integrating the real images and virtual rendering scene, will also be integrated to enhance the application of UAV. The high viewpoint and the combination of real and virtual scene provide engineers augmented views to the construction site. They will increase the possibility of discovering the unfound problems.

Development

We have designed and prototyped an UAV with features important for the construction management: (1) stabilized image capturer, (2) site level image transmission and (3) telecommunication and viewing interface. The stabilized image capturer includes a 3-axis brushless motor gimbal which can continuous adjust the camera to aim a specific direction. The image transmission was carried out by using radio frequency transmitters which specifically optimized by considering the size of the construction site. The interface includes a multi-rotor UAV that has reliable telemetry control, a display that can render the mixed scene between virtual and real views using AR technology, and a wireless local area network (WLAN) system that provides access to realtime image stream for multiple devices.

Tests

From usability tests, we found the system has the potential for discovering unnoticed detail problems in real construction site, such as vehicle path planning conflicts and structure spatial dilemmas. In the near future, we plan to test the UAV-AR solution in three common construction scenarios: typical building site, bridge inspection, and urban planning.

Concept

fastBuild presents a strategy for digital terrain model (DTM) reconstruction using unmanned aerial vehicle (UAV), survey, and image-based modeling technologies. Building a terrain model has been considered a critical issue since it can be used to study terrain transition, flooding simulation, field survey, etc. Existing method uses position data collected by LIDAR (Light Detection and Ranging) to build DTM. However collecting position data by LIDAR is an expensive and time-consuming approach because the surveyor needs to enter the target area with lots of costly instruments no matter how precipitous it is. Thus, we proposed a strategy of using UAV vision and total station survey data to generate imaged-based model since UAV has advantages on flexibility and cost comparing to LIDAR. We have done a field test on a debris fan, which locates in the intersection of Laonong river and Pu-tun-pu-nas tributary, Kaohsiung, Taiwan.

Field Works

We divided the debris into several parts and put 6 to 8 marks as ground control points (GCP). Second, we used UAV equipped with digital camera to photograph each chunk from 300 meters height. At last, we do the post-processing of image-based modeling using PhotoScan, a software that can analyze the aligned photos and build the DTM. It aligns the aerial photos, produces point clouds and exports the DTM.

Results

We compared the result with the one from LIDAR, our strategy is relatively a low-cost and time-saving way to build the DTM. We also improve the precision from 1m to 0.1m. Results of this study suggest a more efficient and applicable approach for DTM reconstruction.

Keywords: Game-based Learning, Interactive Game, Flood Defense, Education.

Flood Game is a tower-defense-like game including multiple flood disaster mitigation methods. Players can learn different flood prevention approaches through the game by choosing methods to protect important city infrastructures. The game begins with a river and some infrastructures along the river. Once hit by a flood, players must choose some engineering methods to protect the infrastructures so that they can keep generating money and the Happy Index can remain high to reflect social stability. Flood Game aim to provoke thought and discussion among students and hopefully other members of society. The games touch upon a range of water topics, which will be enhanced through educational material in both digital and non-digital format, also produced by the team. [Website]

Publication

- M. H. Tsai, M. C. Wen, Y. L. Chang and S. C. Kang (2015, Nov). Game-Based Education for Disaster Prevention. AI & Society, 30(4), 463-475.
- M. C. Wen, M. H. Tsai, S. C. Kang and Y. L. Chang (2013, Oct). Experience disaster mitigation methodology through game playing. Proceeding of the 9th APRU Research Symposium on Multi-Hazards around the Pacific Rim, Taipei, Taiwan.
- M. C. Wen, M. H. Tsai, S. C. Kang and Y. L. Chang (2013, Oct). Flood Game: An Alternative Approach for Disaster Education. Proceedings of 13th International Conference on Construction Applications of Virtual Reality (CONVR), London, UK.
- 蔡孟涵、溫明璋、康仕仲、張玉連 (2013年12月)。創新防災教育學習法：從遊戲中學習。教育部102年校園災害防救實務暨理論研討會，台北，台灣。

Keywords: VR, UAV, Photo Stitch, Web-based, Stereoscopic.

highVR provides a VR tour experience but with high altitude perspective. It is optimized for html5 which can be played on most of modern web browsers. No software or app is needed. Android and iOS devices are also supported.

Demo - Hualien County, Taiwan.
[Try highVR with Desktop!]
[Try highVR with Cardboard Glass!]

Keywords: Risk Analysis, Nuclear Power Plant, Data Visualization, Decision making.

This project is developing a visualization tool for illustrating the risk of a nuclear power plant during different levels of seismic event. We integrated the failure logic tree and the corresponding risk of each components in the tree into one piece. Users are allowed to navigate through 8 different levels of seismic events. A zoomable rectangle grid method is used as a visualized integration outcome. In order to enhance the compatibility of our software, we decide to develop with D3.js under html structure as a web-based software which can be easily deployed to the most of platform. We also optimize the software for mobile devices to raise the mobility and productivity when working in the field. The result shows a great improvement in comprehensibility of the original data. Users can access to the data they need more efficiently and clearly. For future work, the integration of mathematical method behind the risk analysis is needed for equipping the software the ability of automated generation. [Website] [Try Ranal!]

Keywords: 3D, Image Processing, Cinematography, Motion Tracking, Production.

3DMT is developed as a standard operating procedures that adds virtual object into captured footage and composited seamlessly. 3DMT determines the requirements of outputting (realistic) compositions including keying scene setup, lighting condition, character movements, camera configuration and operation, post processing of footage (tracking, grading, etc.), and publishing. The outcome of following 3DMT procedure is promising.