This project builds a mobile computing platform with 3D user interface based on Augmented Reality. The approach combines 3 core technologies, namely Inertial Measurement, Computer Vision and Differential Global Positioning System (GPS) Carrier Phase techniques, into a mobile hybrid tracking system. Currently, prototype Inertial Measurement Units have been constructed and calibrated with new methods that do not require special calibration equipment. A novel GPS real time precision tracking method is has been developed and tested. Possible applications include precision navigation, mobile information delivery and 3D visualization in fields such as design and architecture.

This project develops a rehabilitation system using AR technology to recover hand movements which are impaired due to muscular atrophy after stroke or accidents. The system is a low-cost, portable and fully controlled rehabilitation system which allows users to carry out rehabilitating practice at home. A preliminary working prototype has been developed to track the movements of the hands using data gloves and provide the feedbacks according to the performance of the patients. A prototype based on MEMS sensors to detect upper limb movements has been developed.

This project develops an AR-based collaborative product design environment for multi-users. Product information, e.g., feature information and design history, can be provided to facilitate the users’ tasks. A collaboration framework and an intuitive interface have been developed to facilitate collaborative product design. A novel annotation layout algorithm is developed to avoid overlaps. This AR-CD system can be used for product design and discussion between multiple distributed or co-located users. With the annotating module, this system can be used to introduce new products to potential customers.

ARCNC is an in situ CNC simulation and inspection environment developed using AR technology. In this environment, a machining simulation can be achieved between a virtual workpiece and the real cutter. A prototype system has been developed, and experiments have been conducted on a real 3-axis machine to show the applicability of the system. The reliability of the proposed methods and algorithms, e.g., the hybrid tracking method, the NC code-adaptive cutter registration method, and the MRR-based cutting force estimation model, has been demonstrated. This research has shown for the first time an integration of AR technology into a real CNC machining environment.

This research targets at developing an object manipulation assistance system. The system will detect the user’s activities, e.g., hand movements, and render instructions accordingly to assist the object manipulation. The system can be used in both manufacturing industries and home-based rehabilitation systems. A working prototype has been developed, using RFID, computer vision, and sensor technologies.

In this research, a navigation assistance system is being developed targeting at providing supportive instructions to visually impaired users in both indoor and outdoor environment. A prototype system has been set up, using computer vision, wireless communication, and sensor technology. Indoor experiments have been carried out to prove the concept. Current research focus is to replace the inertial sensor applied in the prototype system with artificial intelligence technology, followed by outdoor experiments.

CNC machining simulation using virtual workpieces and real cutting tools and machining environment allows users to better visualize the machining conditions and to avoid tool collision with fixture and other components.

cutcake1-ysn9a7

This work allows human experts to interact with the real environment to obtain a good layout quickly through an AR interface. Tradeoffs can be made to obtain an optimized arrangement.

(source: Augmented Reality for Facilitating Layout Design, Y.L. Poh, A.Y.C. Nee, S.K. Ong and K. Youcef-Toumi, 2006, Singapore-MIT Alliance Symposium, 17-18 January 2006, Singapore.)

AR-assisted assembly of complex parts allows the assembly sequence to be optimized and provides real-time guidance in training assembly operators, as well as providing them with a ‘feel of the goodness’ of the parts during assembly.

(source: The Virtual Interaction Panel: An Easy Control Tool in Augmented Reality Systems, M.L. Yuan, S.K. Ong and A.Y.C. Nee, 2004, Computer Animation and Virtual Worlds Journal, Special Issue: The Very Best Papers from CASA 2004, Vol. 15, No. 3-4, 425-432.)

(source: Methodologies For Immersive Robot Programming In An Augmented Reality Environment, J.W.S. Chong, S.K. Ong and A.Y.C. Nee, 2007, International Journal of Virtual Reality, Vol. 6, No. 1, 69-79.)

Image registration which is to find the correspondences of two different images acquired from the same scene or object at different times, or from different perspectives, is the key technology to augmented reality. This project attempts to exceed the record of GLOH using a hybrid algorithm which combines the interest point based detectors with region detectors. The algorithm is tested on the Affine Covariant Regions Datasets. The experimental results demonstrate that 97.2% reconstruction similarity can be achieved using our method, 2% higher than any single detectors. This highly a accurate registration algorithm that can be used not only in the area of augmented reality, but also in tampered image detection for forensic investigation.

The goal of this project is to provide an intuitive Human-Robot Interface to facilitate the robot programming process in the real environment. A scaled virtual robot is modeled to duplicate the link/joint configurations of a real robot, and the users are able to “teach” the virtual robot to achieve the basic robotic operations, such as “pick-and-place” operations and path tracking operations.

The objective of the project is to develop an AR-based platform, where human experts can interact with the real environment to obtain an optimized layout arrangement. Based on a camera-based system, the platform is able to construct an AR environment on-site. ARLAD provides virtual objects rendering, collision detection between real and virtual objects and hand manipulation. Each alternative layout pattern will be evaluated based on different criteria. The intuitiveness experienced while working on the platform could provide the users with on-site feedback, which is the most valuable information for layout design. Potential applications are in the field of facility layout, interior design and architecture.

This project develops an AR-based system capable of performing product design and assembly planning in real-time. In this system, the users can view, touch and operate a future product before its physical implementation with lower cost and effort. A bare-hand interaction tool is being developed for natural human-computer interaction.

The potential applications of this project include assisting designers to design products for assembly, facilitating the assembly planning and evaluating the assembly process in the early design stage.

ARMS applies AR to facilitate remote monitoring. Being a peer to peer system, ARMS connects two peers and supports real time information transmission. With ARMS, experts can monitor remote workers by augmenting vivid virtual information in real environment. To facilitate communication between experts and customers, ARMS will offer bidirectional interactions, allowing experts to feel like working in the real environment by manipulating virtual tools with bare hands. The system is suitable for remote maintenance, medical instruction, military applications , etc.

Virtual arrow is augmented in real environment

IDEAS’s objective is to design a 3D Design Space using Augmented Reality to enhance the current design generation process using desktop. Current desktop modeling software uses 2D interfaces to create 3D models, whereas IDEAS will allow direct 3D manipulation and interaction, creating an interactive and intuitive method for 3D modeling. Currently, a conceptual model of IDEAS has been thought out. IDEAS can be used for computer aided design, industrial design and 3D modeling and animations. As it is also intuitive, it may also be used to capture consumer-inspired designs.

An assistive device, a mulit-sensory show, with a feedback system aiding the visually impaired to detect obstacles ahead of them and to determine the reflectance of the surface they are standing on has been developed, allowing the users to have a better sense of their surrounding and instill greater confidence in them.

AR Classroom is an application that presents educational or informative materials in a collaborative Augmented Reality (AR) environment. The AR environment consists of the user’s real surrounding, with computer generated graphics superimposed over it. The user will thus view his or her surrounding together with the computer graphics and be able to interact with the virtual objects in relation to the real world objects.

SmARtWorld aims to enhance the reality of the world around us by giving users context-sensitive information to real life objects.