1,721,046 research outputs found
Anticipation as a strategy: A design paradigm for robotics
No full textAnticipation plays a crucial role during any action, particularly in agents operating in open, complex and dynamic environments. In this paper we consider the role of anticipation as a strategy from a design perspective. Anticipation is a crucial skill in sporting games like soccer, tennis and cricket. We explore the role of anticipation in robot soccer matches in the context of reaching the RoboCup vision to develop a robot soccer team capable of defeating the FIFA World Champions in 2050. Anticipation in soccer can be planned or emergent but whether planned or emergent, anticipation can be designed. Two key obstacles stand in the way of developing more anticipatory robot systems; an impoverished understanding of the "anticipation" process/capability and a lack of know-how in the design of anticipatory systems. Several teams at RoboCup have developed remarkable preemptive behaviors. The CMU Dive and UTS Dodge are two compelling examples. In this paper we take steps towards designing robots that can adopt anticipatory behaviors by proposing an innovative model of anticipation as a strategy that specifies the key characteristics of anticipation behaviors to be developed. The model can drive the design of autonomous systems by providing a means to explore and to represent anticipation requirements. Our approach is to analyze anticipation as a strategy and then to use the insights obtained to design a reference model that can be used to specify a set of anticipatory requirements for guiding an autonomous robot soccer system. © 2010 Springer-Verlag Berlin Heidelberg
Improving human-machine interaction for a powered wheelchair driver by using variable-switches and sensors that reduce wheelchair-veer
No full textTask programming methodology for powered wheelchairs
No full textThe work described in this paper is directed towards applying task oriented methodologies to the creation of new types of powered wheelchairs. The new work has required the creation of a new type of task machine. In addition, new techniques to analyse situations and implement the results within a new user interface have been investigated. This paper describes task machinery in general and the stages required to create a task machine
Fuzzy Sliding Mode Control of Onboard Power Electronics for Fuel Cell Electric Vehicles
Full text linkIn this study, the designing of an intelligent robust control system for onboard power electronic converter, which is utilized in fuel cell electric vehicle, is concerned. It employs sliding-mode control that integrates a fuzzy tuning approach. The DC-DC converters which are employed for the integration of fuel cell power source can regulate the input voltage and provide the output power with high efficiency. Among them, the buck-boost converters with non-inverting property are offered to gain buck and boost features. The proposed control strategy consists of fuzzy logic and sliding mode control methods to synthesize the gains of both controllers. To achieve this goal, a nonlinear average model of the converter is extracted. Then a fuzzy sliding mode controller is planned to adjust the output voltage of the converter. Furthermore, robustness and stability of the controller are proved by Lyapunov theory and they are fulfilled completely. The obtained computer results demonstrate the ability of the control policy throughout diverse situations
Dempster-Shafer Credal Probabilistic Circuits
No full textProbabilistic circuits are deep, tractable generative models capable of computing various types of exact inferences. However, their traditional specifications do not fully account for epistemic uncertainty. To address this, credal probabilistic circuits were introduced, incorporating a way to manage such uncertainty. We propose a novel framework for learning the structure and parameters of credal probabilistic circuits, leveraging the Dempster-Shafer theory of evidence. Unlike previous credal approaches, the framework handles both discrete and continuous data and allows for the use of multiple classification criteria. We conclude by presenting some preliminary experimental results, demonstrating the performance of the proposed models compared to commonly used probabilistic circuits across a range of classification tasks
Knowledge-based expert system using a set of rules to assist a tele-operated mobile robot
Full text linkIntegration of fuzzy c-means and artificial neural network for short-term localized rainfall forecast in tropical climate
No full textA Belief Rule-Based Expert System to Assess Bronchiolitis Suspicion from Signs and Symptoms Under Uncertainty
No full textBronchiolitis is a common disease in children and an acute viral infection of the bronchioles that affects millions of children around the world. The assessment of the suspicion of this disease is usually carried out by looking at its signs andsymptoms. However, these signs and symptoms cannot be measured with cent percent certainty, resulting in inaccuracy in determining the occurrence of Bronchiolitis. Therefore, this paper presents the development of a Belief Rule-Based Expert System (BRBES) to assess the suspicion of Bronchiolitis by usingsigns and symptoms under uncertainty. The recently developed generic belief rule-based inference methodology by using evidential reasoning (RIMER) acts as the inference engine of this BRBES while belief rule base as the knowledge representation schema. The knowledge base of the system is constructed byusing real patient data and expert opinion from Bangladesh. It has been observed that the results generated by using the BREBES are more reliable than those from the manual system
Person Detection in Thermal Videos Using YOLO
No full textIn this paper, the task of automatic person detection in thermal images using convolutional neural network-based models originally intended for detection in RGB images is investigated. The performance of the standard YOLOv3 model is compared with a custom trained model on a dataset of thermal images extracted from videos recorded at night in clear weather, rain and fog, at different ranges and with different types of movement – running, walking and sneaking. The experiments show excellent results in terms of average precision for all tested scenarios, and a significant improvement of performance for person detection in thermal imaging with a modest training set
- …
