International Journal of Information Acquisition

In this paper, a four layer neuro-fuzzy architecture of multi-sensor fusion is developed for a fault diagnosis system which is applied to an industrial fish cutting machine. An important characteristic of the fault diagnosis approach developed in this paper is to make an accurate decision of the machine condition by fusing information acquired from three types of sensors: Accelerometer, microphone and charge-coupled device (CCD) camera. Feature vectors for vibration and sound signals from their fast Fourier transform (FFT) frequency spectra are defined and extracted from the acquired information. A feature-based vision method is applied for object tracking in the machine, to detect and track the fish moving on the conveyor. A four-layer neural network including a fuzzy hidden layer is developed in the paper to analyze and diagnose existing faults. Feature vectors of vibration, sound and vision are provided as inputs to the neuro-fuzzy network for fault detection and diagnosis. By proper training of the neural network using data samples for typical faults, six crucial faults in the fish cutting machine are detected with high reliability and robustness. On this basis, not only the condition of the machine can be determined for possible retuning and maintenance, but also alarms to warn about impending faults may be generated during the machine operation.

This paper presents the architecture used to develop a micro-robot for narrow pipes inspection. Both the electromechanical design and the control scheme will be described. In pipe environments it is very useful to have a method to retrieve information of the state of the inner part of the pipes in order to detect damages, leaks and holes. Due to the different types of existing pipes, a modular approach with different types of modules has been chosen in order to be able to adapt to the shape of the pipe and to chose the most appropriate gait. The micro-robot has been designed for narrow pipes, a field in which there are not many prototypes. The micro-robot incorporates a camera module for visual inspection and several drive modules for locomotion and turn (helicoidal, inchworm, two degrees of freedom rotation). The control scheme is based on semi-distributed behavior control and is also described. A simulation environment is also presented for prototype testing.

The common robot communication platform using Web Service is well accepted and becoming popular gradually. One of the key components for robot communication platform is the reliable messaging which provides reliable high performance message transfer. Both conforming to standard specifications and interoperability among multiple implementations are critical requirements for the platform, because different robots and different services should be able to connect and communicate each other on the platform. However, there is no conformance/interoperability testing tool for the reliable message component of Web Services. This paper describes requirements for the conformance and interoperability testing for Web Service technologies, and how we developed the verification suit that satisfies the requirements by automated error case verification model. This paper also reports the interoperability verification results for emerging reliable message components of Web Services and our contribution to the international standardization group based on this verification results.

Gymnotiform swimmer uses single undulating anal fin below the rigid body as the propulsor. This special morphological structure is generally thought to have advantages in reducing the friction drag in undulatory movements. The finned body form and the undulatory swimming gait give the inspiration to the design of underwater robot. We have developed a robotic fish (Nanyang KnifeFish, NKF-I), which learns from a Gymnotiform swimmer: Black Ghost Knifefish. A new version of biomimetic robotic fish (NKF-II) is presented in this paper. To mimic the actual flexible fin of real fish, an optimized fin-like propulsion mechanism is modelled with a series of connecting linkages for enhancing its propulsive performance. Workspace of the proposed mechanism is studied and optimized. Swimming motion function is also established for the planning of the undulatory gait movement. The manoeuvrability and buoyancy control has also been achieved by the integration of mechatronics design of a buoyancy module with the undulating fin. The design philosophy of modularity has also been incorporated into the design of the fish. Hence, the modularity of the prototypes not only provides versatility for the robot system and also facilitates in the maintenance and future development of fish with undulating fins/body, as new modules can be added to replace the existing modules easily. By reconfiguring the components of NKF-II, other version of robotic fish with undulating fins/body is derived. This evolution of the undulating fins is also discussed.

Driving is a challenging task, especially for those with color vision deficiencies. Intelligent Transport Systems (ITS) can provide useful information and make driving safe and convenient. A real time traffic light recognition system is presented in this paper for improving public safety and facilitating color deficient drivers. It may also be included as a part of ITS. The presented system consists of a digital video camera to record traffic lights and a portable PC to process images in real time. It uses various techniques of color detection, feature matching with normalized cross-correlation (NCC), and mathematical shape analysis for the traffic light recognition. For color detection, we obtained an initial solution by using RGB component adjustment, thresholding algorithm, and median filter. In dealing with illumination changes with weather and time, a simple adaptation method was developed. Feature matching with NCC was used after color detection to further detect and recognize the traffic lights. To improve the system's tolerance and robustness, a mathematical shape analysis was undertaken to obtain the final results. Numerous experiments were conducted to demonstrate the effectiveness and practicability of the system with images under different weather conditions. The average recognition ratio is higher than 95% from the testing results. The average processing time is 30 ms per frame, making the system suitable in real time conditions. Audio alert is added to the current system as an integral part of a portable system to be developed.

A position management system for forklifts involved in a warehouse storage task is proposed. With a simple, low-cost laser beam sensor and reflective stickers, the position of the forklift can be established. Furthermore, by combining probability estimation algorithms, the possible location of forklifts can be managed, and the accuracy of positioning can be improved through the exchange of information on a sensor network. Finally, the layout, which has a better system performance, is also discussed and proposed. With this system, forklift operations are easier, and the cost of a warehouse navigation system decreases.

In this paper, the author presents an approach for automated knowledge extraction from rise time auto-correlated patterns by using self-organizing maps and k-means clustering. The extracted knowledge in terms of rules will be used as knowledge base for an expert system. Rise-time auto-correlated data patterns are used as a learning data set. The produced knowledge based was verified by using a conventional expert system.