In this paper, we analyze a sparse framework for hybrid radiolocalization of multiple emitters in the presence of a simple multipath channel. The scenario under consideration includes multiple non-collaborative target...
详细信息
Real-time full-body tracking in VR is important for providing realistic experiences, especially for applications such as training, education, and social VR. The Microsoft Kinect v2 sensor can provide skeleton data for...
详细信息
The proceedings contain 17 papers. The special focus in this conference is on applications in Computational Intelligence. The topics include: Comparison of evolutionary algorithms for estimation of parameters of the e...
ISBN:
(纸本)9783030030223
The proceedings contain 17 papers. The special focus in this conference is on applications in Computational Intelligence. The topics include: Comparison of evolutionary algorithms for estimation of parameters of the equivalent circuit of an AC motor;Cost-balance setting of MapReduce and spark-based architectures for SVM;nonlinear methodologies for climate studies in the Peruvian Northeast Coast;spectral image fusion for increasing the spatio-spectral resolution through side information;evaluation of a modified current selective harmonic elimination technique applied in low voltage power systems;about the effectiveness of teleconsults to evaluate the progress of type-2 diabetes and depression;optimal dimensioning of electrical distribution networks considering stochastic load demand and voltage levels;application of transfer learning for object recognition using convolutional neural networks;SOM-like neural network and differential evolution for multi-level image segmentation and classification in slit-lamp images;Implementation of a neural control system based on PI control for a non-linear process;filter banks as proposal in electrical motors fault discrimination;discrimination of nonlinear loads in electric energy generation systems using harmonic information;a systematic literature review of hardware neural networks;on computing the variance of a fuzzy number;eHealth services based on Monte Carlo algorithms to anticipate and lessen the progress of type-2 diabetes.
This book includes papers from the 5th International conference on Robot Intelligence Technologyand applications held at KAIST, Daejeon, Korea on December 1315, 2017. It covers the following areas: artificial intellig...
ISBN:
(纸本)9783319784519
This book includes papers from the 5th International conference on Robot Intelligence Technologyand applications held at KAIST, Daejeon, Korea on December 1315, 2017. It covers the following areas: artificial intelligence, autonomous robot navigation, intelligent robot system design, intelligent sensing and control, and machine vision. The topics included in this book are deep learning, deep neural networks, image understanding, natural language processing, speech/voice/text recognition, reasoning & inference, sensor integration/fusion/perception, multisensor data fusion, navigation/SLAM/localization, distributed intelligent algorithms and techniques, ubiquitous computing, digital creatures, intelligent agents, computer vision, virtual/augmented reality, surveillance, pattern recognition, gesture recognition, fingerprint recognition, animation and virtual characters, and emerging applications. This book is a valuable resource for robotics scientists, computer scientists, artificial intelligence researchers and professionals in universities, research institutes and laboratories.
The Automated Global Feature Analyzer (TM) (AGFA (TM)) is a generically applicable automated sensor-data-fusion, feature extraction, feature vector clustering, anomaly detection, and target prioritization framework. A...
详细信息
ISBN:
(数字)9781510617902
ISBN:
(纸本)9781510617902
The Automated Global Feature Analyzer (TM) (AGFA (TM)) is a generically applicable automated sensor-data-fusion, feature extraction, feature vector clustering, anomaly detection, and target prioritization framework. AGFA (TM) operates in the respective feature space delivered by the sensor(s). In this paper we provide an overview of the inner workings of AGFA (TM) and apply AGFA (TM) to planetary imagery, representative of past, current, and future planetary missions, to demonstrate its automated and objective (i.e., unbiased) anomaly detection and target prioritization (i.e., region-of interest delineation) capabilities. Imaged operational areas are locally processed via a cascade of image segmentation, visual and geometric feature extraction, agglomerative clustering, and principal components analysis. Resulting clusters are labeled based on relative size and location in feature space. Anomalous regions may be considered immediate targets for follow-up in-situ investigation by local robotic agents, which can be directed via autonomous telecommanding, e.g., as part of a Tier-Scalable Reconnaissance mission architecture. These capabilities will be essential for driving fully autonomous (CISR)-I-4 missions of the future, since the speed of light prohibits "real time" Earth-controlled conduct of planetary exploration beyond the Moon.
The advent of the Internet-of-Things has introduced a new paradigm that supports a decentralized and hierarchical communication architecture, where a great deal of analytics processing occurs at the edge and at the en...
详细信息
ISBN:
(纸本)9781509049400
The advent of the Internet-of-Things has introduced a new paradigm that supports a decentralized and hierarchical communication architecture, where a great deal of analytics processing occurs at the edge and at the end-devices instead of in the Cloud. To map the embedded-systems requirements, we present a holistic research approach to the development of low-power architectures inspired by the human brain, where process development and integration, circuit design, system architecture, and learning algorithms are simultaneously optimized. This paper is organized as follows: We begin with a survey of recent research on the human brain and a historical perspective of cognitive neuroscience. Then, artificial intelligence is introduced, and the challenges of Deep Learning systems (in terms of power requirements) are addressed. The key reasons to distribute intelligence over the whole network are discussed. To emphasize the need for low-power solutions, a quantitative benchmark of existing specialized edge platforms that can execute machine-learning algorithms on conventional embedded hardware is presented. The primary focus of this paper will be on the implementation of optimized neuromorphic hardware as a highly promising solution for future ultra-low-power cognitive systems. We show that emerging technologies (such as advanced CMOS, 3D technologies, emerging resistive memories, and Silicon photonics), coupled with novel brain-inspired paradigms, such as spike-coding and spike-time-dependent-plasticity, have extraordinary potential to provide intelligent features in hardware, approaching the way knowledge is created and processed in the human brain. Finally, we conclude with our vision of the enabled future disruptive applications and a discussion of the main challenges which should be tackled to exploit the full potential of brain-inspired technologies.
An Attitude and Heading Reference System (AHRS) comprising accelerometers, gyroscopes and magnetometers can provide roll, pitch and heading information. AHRS is utilized in many applications such as navigation, augmen...
详细信息
ISBN:
(纸本)9781538673928
An Attitude and Heading Reference System (AHRS) comprising accelerometers, gyroscopes and magnetometers can provide roll, pitch and heading information. AHRS is utilized in many applications such as navigation, augmented/virtual reality, and mobile mapping. The AHRS mechanization involves integration of angular rate measurement to provide high rate orientation but with unbounded drifts due to accumulation of random noise. To reduce drifts, mechanization output is combined with absolute measurement from magnetometer and accelerometer using Extended Kalman Filter(EKF). EKF accuracy is greatly affected by process covariance matrix (Q) and measurement noise covariance matrix(R). Conventional stochastic modeling approaches to determine Q and R parameters do not guarantee best performance. This paper proposes a systematic procedure for EKF parameters optimization using a hybrid statistical and genetic algorithms (GA) approach. The proposed approach has been verified on real data collected by an inertial measurement unit. Results showed that the Q and R can be optimized within few GA iterations outperforming conventional EKF parameter estimation methods.
Driven by applications like Micro Aerial Vehicles (MAVs), driver-less cars, etc, localization solution has become an active research topic in the past decade. In recent years, Ultra Wideband (UWB) emerged as a promisi...
详细信息
ISBN:
(纸本)9781538660898
Driven by applications like Micro Aerial Vehicles (MAVs), driver-less cars, etc, localization solution has become an active research topic in the past decade. In recent years, Ultra Wideband (UWB) emerged as a promising technology because of its impressive performance in both indoor and outdoor positioning. But algorithms relying only on UWB sensor usually result in high latency and low bandwidth, which is undesirable in some situations such as controlling a MAV. To alleviate this problem, an Extended Kalman Filter (EKF) based algorithm is proposed to fuse the Inertial Measurement Unit (IMU) and UWB, which achieved 80Hz 3D localization with the significantly improved accuracy and almost no delay. To verify the effectiveness and reliability of the proposed approach, a swarm of 6 MAVs is set up to perform a light show in an indoor exhibition hall.
Identifying and detecting glass is an important part of robotic safety at office conditions. Existing glass detection algorithms rely on high-cost lidars or ultrasonic rings. We propose an efficient and simple algorit...
详细信息
ISBN:
(纸本)9781728103778
Identifying and detecting glass is an important part of robotic safety at office conditions. Existing glass detection algorithms rely on high-cost lidars or ultrasonic rings. We propose an efficient and simple algorithm to detect glass based on sensorfusion of ultrasound sensors and laser scanners data that can adapts to all kinds of lidars. Combine the glass detection results with gmapping algorithm to generate a more accurate map, the maximum connected domain algorithm used to remove noise from the map. We tested the glass detection algorithm on the robot platform. Experiments showed that about 93% of the glass can be detected correctly.
We present a simple approach for sensor registration in target tracking applications. The proposed method uses targets of opportunity and, without making assumptions on their dynamical models, allows simultaneous cali...
详细信息
We present a simple approach for sensor registration in target tracking applications. The proposed method uses targets of opportunity and, without making assumptions on their dynamical models, allows simultaneous calibration of multiple three- and two-dimensional sensors. Whereas for two-sensor scenarios only relative registration is possible, in practical cases with three or more sensors unambiguous absolute calibration may be achieved. The derived algorithms are straightforward to implement and do not require tuning of parameters. The performance of the algorithms is tested in a numerical study.
暂无评论