The RAPPID Project: Symbiosis between Industrial Requirements and MAS Research

The RAPPID project (Responsible Agents for Product-Process Integrated Design) is developing agent-based software tools and methods that use market place dynamics among members of a distributed design team to coordinate set-based design of discrete manufactured products. This report describes the interplay of industrial requirements and Multi-Agent System (MAS) research in the design, implementation, and testing of RAPPID. Like any industrial project, RAPPID begins with the requirements of the problem domain, and draws selectively from research results to meet those requirements. However, the flow of information is not unidirectional. In the process of addressing its requirements, RAPPID developed some new concepts that hold promise for broader application to distributed constraint optimization. RAPPID is work in process, not a completed product, and this report includes an assessment of what needs to be done in addition to what has been accomplished.     

Industrial actuator benchmark for fault detection and isolation

Feedback control systems are vulnerable to faults within the control loop, because feedback actions may cause abrupt responses and process damage when faults occur. Such faults can be detected by model-based methods for fault detection and isolation (FDI) but research results have not been widely accepted in industry. One reason has been a scarcity of realistic examples for testing FDI methods against industrial systems. These special section papers focus on a common benchmark example, an electro-mechanical position servo, used in speed control of large diesel engines. The result is a platform for comparison of FDI methods and a gathering together of design experience on a simple, yet very realistic, industrial example. This paper introduces the benchmark problem, overviews the FDI methods used within the papers and discusses the results.     

'Toy problem' as the benchmark test for teleoperation systems

A unified hand/arm master–slave system was developed to conduct dexterous tasks by teleoperation. Both the master and the slave consist of a two-fingered hand and a 3-d.o.f. planer manipulator. The system is also aimed at investigating the bottleneck of autonomous robotic manipulation systems. To evaluate the performance of individual teleoperation systems all over the world under equal conditions, we propose toy block assembling (LEGO® blocks) as the benchmark test for teleoperation systems. Using our hand/arm master-slave system, we announce the measured completion time for assembling some block structures, expecting other research group to try the same task in the future.     

Vision at work visual defect and the visual demand of tasks

Several related studies of visual tasks in a large communications undertaking have revealed much defect of vision and many sources of visual fatigue and inefficiency. The nature of visual fatigue is described briefly and a programme to facilitate visual tasks is outlined.     

Analytical and numerical solutions for a reliability-based benchmark example

The aim of this elementary reliability-based truss topology example is to serve as a benchmark for checking on the validity, accuracy and convergence of FE-based numerical topology optimization methods. The above problem has been solved analytically by using an extension of the optimal layout theory (Prager and Rozvany) and the solution has been verified numerically by a first order reliability approach (FORM) combined with a material distribution method (SIMP).     

Vision and navigation for the Carnegie-Mellon Navlab

A distributed architecture articulated around the CODGER (communication database with geometric reasoning) knowledge database is described for a mobile robot system that includes both perception and navigation tools. Results are described for vision and navigation tests using a mobile testbed that integrates perception and navigation capabilities that are based on two types of vision algorithms: color vision for road following, and 3-D vision for obstacle detection and avoidance. The perception modules are integrated into a system that allows the vehicle to drive continuously in an actual outdoor environment. The resulting system is able to navigate continuously on roads while avoiding obstacles     

Integrating face and gait for human recognition at a distance in video.

This paper introduces a new video-based recognition method to recognize noncooperating individuals at a distance in video who expose side views to the camera. Information from two biometrics sources, side face and gait, is utilized and integrated for recognition. For side face, an enhanced side-face image (ESFI), a higher resolution image compared with the image directly obtained from a single video frame, is constructed, which integrates face information from multiple video frames. For gait, the gait energy image (GEI), a spatio-temporal compact representation of gait in video, is used to characterize human-walking properties. The features of face and gait are obtained separately using the principal component analysis and multiple discriminant analysis combined method from ESFI and GEI, respectively. They are then integrated at the match score level by using different fusion strategies. The approach is tested on a database of video sequences, corresponding to 45 people, which are collected over seven months. The different fusion methods are compared and analyzed. The experimental results show that: 1) the idea of constructing ESFI from multiple frames is promising for human recognition in video, and better face features are extracted from ESFI compared to those from the original side-face images (OSFIs); 2) the synchronization of face and gait is not necessary for face template ESFI and gait template GEI; the synthetic match scores combine information from them; and 3) an integrated information from side face and gait is effective for human recognition in video.     

Output regulation for the TORA benchmark via rotational position feedback

The present work deals with the design of a feedback controller that solves an output regulation problem for the nonlinear benchmark example known as TORA. The controller uses only measurements of the rotational position and is obtained through a certainty-equivalence approach.     

Active disturbance rejection control for the ALSTOM gasifier benchmark problem

A new control solution based on a unique active disturbance rejection control (ADRC) paradigm is proposed for the ALSTOM gasifier control problem. In the new control method, the disturbances, including unknown dynamics and external disturbances to the system, are treated as an augmented state that is estimated and then eliminated in real-time. A first-order ADRC scheme for the ALSTOM gasifier is designed. The simulation results show that the new control method is superior to the PI solution by Simm and Xue in terms of the integral of the absolute error of gas pressure disturbance tests at 0% load, the no-overshoot peak value of the gas temperature and the short time for the bed mass to return to the set point during a ramping test, and the ability to reject large coal quality disturbances.     

Multiple hypothesis-testing approach to FDI for the industrial actuator benchmark

Detection of current faults in a D.C. motor with unknown load torques may not be feasible with linear methods and threshold logic. This paper suggests an FDI scheme with a bank of observers where a nonlinear observer is designed using the structure of the current fault. Multiple-hypothesis testing of the observer residuals is applied to discriminate between faults and unknown input. The strategy has been applied to a simulation of the diesel engine actuator benchmark problem with an encouraging progress on performance compared to other results.     

Distributed computing for vision: architecture and a benchmark test

Computer vision algorithms are notorious for their computational expense. Distributed vision, the use of more than one processor, can decrease computation costs and speed up algorithms. There are various ways to do this, ranging from parallelism at the sensor level to true multiprocessor systems. This correspondence first describes a system of the latter type: a system of microprocessors on a high-speed bus. A canonical vision task, locating a number of objects and measuring certain two-dimensional features of those objects, serves as a benchmark test for the system. An algorithm for this task is presented. Performance measures are compared from implementations on the distributed system, a Vax 11/750, and a Vax 11/780. Results indicate that three microprocessors outperform a Vax 11/780 at this task. Finally, other more interesting distributed algorithms are briefly discussed.     

Remember and forget: video and text fusion for video question answering

Video question answering (Video QA) has received much attention in recent years. It can answer questions according to the visual content of a video clip. Video QA task can be solved only according to the video data. But if the video clip has some relevant text information, It can also be solved by using the fused video and text data. How to select the useful region features from the video frames and select the useful text features from the text information needs to be solved. And how to fuse the video and text features also needs to be solved. Therefore, we propose a forget memory network to solve these problems. The forget memory network with video framework can solve Video QA task only according to the video data. It can select the useful region features for the question and forget the irrelevant region features from the video frames. The forget memory network with video and text framework can extract the useful text features and forget the irrelevant text features for the question. And it can fuse the video and text data to solve Video QA task. The fused video and text features can help improve the experimental performance.     

Symbiosis or assimilation: critical reflections on the ontological self at the precipice of Total Data

Contemporary data practices are inducing a convergent saturation point (conceptually referred to as Total Data) wherein every human action, reaction, interaction, transaction, thought or desire is quantified, reified, recorded and used. Physical or virtual, all is recorded, known or unknown, seen or unseen, until data permeates every facet of our shared human existence. The implications of this eventuality are potentially so far reaching that the very notion or concept of who we are might be fundamentally altered, resulting in new ontologies of the self in a world of Total Data. This polemic paper reflects on the implications that Total Data has for the ontological self in a range of individual and shared contexts, and considers the potential it has to ultimately be symbiotic or assimilatory. It suggests that the current trajectory for Total Data is more assimilatory than symbiotic, demonstrating more potential to collectively monitor and control people than to emancipate and empower them. In response, it calls for an authentic debate and reassessment of current data practices, and for an urgent reprioritisation of core and enduring human-centred values and symbiosis in technological systems development to emancipate and empower people living in a Total Data world.     

Efficient compression and network adaptive video coding for distributed video surveillance

The availability of cheap network based video cameras and the prevalence of wireless networks has lead to a major thrust towards deployment of large scale Distributed Video Surveillance (DVS) systems. This has opened up an important area of research to deal with the issues involved in DVS system for efficient collection and transmission of large scale video streams from the cameras at the guarded sites, to the end users in possibly constrained network conditions. In this paper, we propose a framework based on content-based video classification and scalable compression scheme to provide a robust bandwidth efficient video transmission for DVS. The scheme builds on a Discrete Wavelet Transform (DWT) based Color-Set Partitioning for Hierarchical Trees (CSPIHT) coding to obtain a scalable bitstream. Wavelet domain segmentation and compression assists in development of a DVS architecture. The architecture includes a novel module for dynamic allocation of Network bandwidth based on the current available resources and constraints. Different frame constituents are optimally coded based on their relative significance, perceptual quality, and available estimate of network bandwidth. Experimental result over different video sequences and simulations for Network conditions demonstrate the efficient performance of the approach.     

An observer and signal-processing approach to FDI for the industrial actuator benchmark test

In order to apply the best fault-detection and diagnosis scheme, it is required to investigate the process model profoundly and the kinds of faults to be detected. Especially, the process excitation and the effect of the fault being considered play an important role. This is the starting point for the choice of one of the various model-based fault-detection methods. According to this strategy, two different approaches, an observer-based and a signal-based approach, are selected for the two given faults of the benchmark task. It is shown that the use of adaptive thresholds can significantly improve the performance of the fault-detection scheme with respect to the false alarm rate and the delay in detection.