Search

Particle Swarm Optimization (PSO) was first introduced as a concept for a non-linear optimizer by Kennedy and Eberhart in 1995. Their seminal work articulates a technique of evolutionary computation, which has its origin in artificial intelligence and simplified social models such as bird flocking and fish schooling (Kennedy & Eberhart, Nov. 1995; Kennedy & Eberhart, Oct. 1995). Its early appeal lay in its use of only primitive mathematical operators and computational economy with regard to both memory and speed. The authors were influenced by the work of Reynolds, Heppner and Grenander in modeling bird flocking and recognized that the fundamental hypothesis to the development of PSO is that an evolutionary advantage is obtained by the social sharing of information among members of the same species. They stated that the simulation of the graceful but unpredictable choreography of a bird flock by collision-proof agents could be used as an effective optimizer for a wide range of functions.

Multi-step manipulation tasks in unstructured environments are extremely challenging for a robot to learn. Such tasks interlace high-level reasoning that consists of the expected states that can be attained to achieve an overall task and low-level reasoning that decides what actions will yield these states. We propose a model-free deep reinforcement learning method to learn multi-step manipulation tasks. We introduce a Robotic Manipulation Network (RoManNet), which is a vision-based model architecture, to learn the action-value functions and predict manipulation action candidates. We define a Task Progress based Gaussian (TPG) reward function that computes the reward based on actions that lead to successful motion primitives and progress towards the overall task goal. To balance the ratio of exploration/exploitation, we introduce a Loss Adjusted Exploration (LAE) policy that determines actions from the action candidates according to the Boltzmann distribution of loss estimates. We demonstrate the effectiveness of our approach by training RoManNet to learn several challenging multi-step robotic manipulation tasks in both simulation and real-world. Experimental results show that our method outperforms the existing methods and achieves state-of-the-art performance in terms of success rate and action efficiency. The ablation studies show that TPG and LAE are especially beneficial for tasks like multiple block stacking. Code is available at: https://github.com/skumra/romannet, Comment: Accepted for publication in IEEE Robotics and Automation Letters (RA-L)

In this work, we present a deep reinforcement learning based method to solve the problem of robotic grasping using visio-motor feedback. The use of a deep learning based approach reduces the complexity caused by the use of hand-designed features. Our method uses an off-policy reinforcement learning framework to learn the grasping policy. We use the double deep Q-learning framework along with a novel Grasp-Q-Network to output grasp probabilities used to learn grasps that maximize the pick success. We propose a visual servoing mechanism that uses a multi-view camera setup that observes the scene which contains the objects of interest. We performed experiments using a Baxter Gazebo simulated environment as well as on the actual robot. The results show that our proposed method outperforms the baseline Q-learning framework and increases grasping accuracy by adapting a multi-view model in comparison to a single-view model., 6 pages, 6 figures, 2020 IEEE International Conference on Automation Science and Engineering (CASE)

This paper presents an experimentation platform for human robot collaboration as a system of systems as well as proposes a conceptual framework describing the aspects of Human Robot Collaboration. These aspects are Awareness, Intelligence and Compliance of the system. Based on this framework case studies describing experiment setups performed using this platform are discussed. Each experiment highlights the use of the subsystems such as the digital twin, motion capture system, human-physiological monitoring system, data collection system and robot control and interface systems. A highlight of this paper showcases a subsystem with the ability to monitor human physiological feedback during a human robot collaboration task., (Accepted in SOSE 2019)

In this paper, an analysis of the sensing volume coverage of robot workspace as well as the shared human-robot collaborative workspace for various configurations of on-robot Time-of-Flight (ToF) sensor array rings is presented. A methodology for volumetry using octrees to quantify the detection/sensing volume of the sensors is proposed. The change in sensing volume coverage by increasing the number of sensors per ToF sensor array ring and also increasing the number of rings mounted on robot link is also studied. Considerations of maximum ideal volume around the robot workspace that a given ToF sensor array ring placement and orientation setup should cover for safe human robot interaction are presented. The sensing volume coverage measurements in this maximum ideal volume are tabulated and observations on various ToF configurations and their coverage for close and far zones of the robot are determined., Comment: (Draft) Submitted to IEEE SMC 2019

14.1 System of autonomous rovers 366 14.1.1 Stationary sensors and sensor networks 367 14.2 Haptically controlled base robot 367 14.2.1 Electrical and mechanical construction 368 14.2.2 Haptic control: the haptic gravitational field (HGF) 369 14.2.3 Operation in the system of autonomous rovers 370 14.2.3.1 Communication schemes 370 14.2.3.2 Control schemes 371 14.3 Swarm robots 371 14.3.1 Mechanical construction and components 371 14.3.2 Navigation solution with GPS 372 14.3.2.1 The interface 372 14.3.2.2 Understanding the parameters 373 14.3.2.3 Calculating the heading angle 373 14.3.3 Sensor fusion 374 14.4 Application: robust threat monitoring 375 14.4.

A novel Complex-Valued Convolutional Neural Network (CV-CNN) model is proposed for the detection process in Multiple Input Multiple Output next generation wireless networks. The proposed model is used a as an auto-encoder based CV-CNN which is based on using the absolute value as the activation function and the L1-Norm as the loss function instead of the Euclidean norm. The proposed model is used for the joint detection of the index of the transmitted symbols and the index of the spatial channel matrix of the active antennas that are used for the high spectrum efficient Multi-Symbol Generalized Spatial Modulation systems. Simulation results show that the performance of the proposed CNN measured in terms of the Bit Error Rate (BER) is at most 0.5dB less than the performance of the Maximum Likelihood (ML) detector with the advantage of reducing the computational complexity at the receiver. The computational complexity is reduced by 14.71% which is achieved by the reduction of the search space at the output of the maximum pooling layer.

In this paper, a novel self-interference cancellation receiver is proposed to maximize the spectrum utilization for satellite communications. To the best of our knowledge, the proposed system is the first system that performs self-interference cancellation without the estimation of the signal parameters (frequency, phase, and gain offsets) and without the generation of the self-interfering signal. Correlation Receiver is used for the demodulation process followed by a subtracter to cancel the self-interfering in-phase and quadrature phase components at each earth station of a two way Very Small Aperture Terminal (VSAT) satellite communication link. The proposed technique achieves more than 50% saving of the bandwidth of the space segment over the traditional transmission. Simulation results show that the receiver performance in terms of Bit Error Rate (BER) versus the signal to noise ratio (SNR) at the Maximum Likelihood (ML) detector output of each link is almost the same as the case of using two different transmit frequencies. Besides to the reliable performance, the proposed method simplifies the receiver at each earth station.

Objective: The cytotoxic effects of natural rubber gloves and urinary catheters on different cells have been earlier reported. In order to reduce these effects, various coating techniques such as silver or polymer (hydrogel, silicone, polytetrafluoroethylene (PTFE)) coatings have been developed. However, during manufacturing process of silicone-coated rubber catheters, similar to latex catheters, zinc compounds, which have cytotoxic effects are added to the natural rubber base. The aim of this study was to compare the in vitro cytotoxic effects of rubber catheters with the silicone-coated rubber and to investigate the causes of the toxic effect. Method: The cytotoxic effects of total 48 urinary catheters in three different structure (natural rubber, silicone, silicone-coated rubber) and 8 different brands were analysed in mouse connective tissue fibroblast cell culture (L-929) by qualitative (imaging and scoring) and quantitative (MTT) methods. The levels of the elements in catheter extracts were examined in an Atomic Absorption Spectrophotometer (AAS). Results: Compared with standard polyethylene used as a negative control, the extracts from 100% silicone did not change vitality of the fibroblast cells OZET Amac: Dogal kaucuk yapida olan sonda ve cerrahi eldivenlerin farkli hucreler uzerindeki sitotoksik etkileri oldugu bildirilmistir. Bu etkilerin azaltilmasi amaciyla gumus ya da polimerik (hidrojel, silikon, politetrafloroetilen) kaplamalar gibi degisik kaplama teknikleri gelistirilmistir. Kaplama sondalarin uretim asamasinda, kaucuk sondalara benzer sekilde hizlandirici olarak cinko bilesikleri kullanilmaktadir. Bu calismanin amaci, kaucuk sondalarin in vitro sitotoksik etkilerini, silikon kaplanan kaucuk sondalar ile karsilastirmak ve toksisite nedenini

Fig. 2. Proposed method result of different rules for Im 1 ( Δ = 110 , ? = 0.62 ), (a) Original Image, (b) Ground Truth, (c) Rule 47, (d) Rule 170, (e) Rule 367, (f) Rule 510.Display Omitted This study is the application of 2D linear cellular automata (CA) rules with the help of fuzzy membership function to the problems of edge detection.An efficient and simple thresholding technique of edge detection based on CA transition rules optimized by Particle Swarm Optimization method (PSO) is proposed.Results of the proposed to the selected 22 images from the Berkeley Segmentation Dataset (BSDS) are presented.Comparison with some classical Sobel and Canny results is included.Baddeley Delta Metric (BDM) is used for the performance index to compare the results. In this paper we discuss the application of two-dimensional linear cellular automata (CA) rules with the help of fuzzy heuristic membership function to the problems of edge detection in image processing applications. We proposed an efficient and simple thresholding technique of edge detection based on fuzzy cellular automata transition rules optimized by Particle Swarm Optimization method (PSO). Finally, we present some results of the proposed linear rules for edge detection to the selected 22 images from the Berkeley Segmentation Dataset (BSDS) and compare with some classical Sobel and Canny results. Also, Baddeley Delta Metric (BDM) is used for the performance index to compare the results.

The twin rotor MIMO system (TRMS) is a helicopter-like system that is restricted to two degrees of freedom, pitch and yaw. It is a complicated nonlinear, coupled, MIMO system used for system identification, the verification of control methods and observers. This paper details the design procedure for a suboptimal tracking controller using a linear quadratic regulator (LQR) with integral action. It was found that the LQR controller with integral action (LQI) provided performance superior to existing optimal controllers in the literature.

In this paper we discuss the application of two-dimensional linear cellular automata (CA) rules with the help of fuzzy heuristic membership function to the problems of edge detection in image processing applications. We proposed an efficient and simple thresholding technique of edge detection based on fuzzy cellular automata transition rules optimized by Particle Swarm Optimization method (PSO). Finally, we present some results of the proposed linear rules for edge detection and compare with some classical results.

This research proposes a Brain Computer Interface as an interactive and intelligent Image Search and Retrieval tool that allows users, disabled or otherwise to browse and search for images using brain signals. The proposed BCI system implements decoding the brain state by using a non-invasive electroencephalography (EEG) signals, in combination with machine learning, artificial intelligence and automatic content and similarity analysis of images. The user can spell search queries using a mental typewriter (Hex-O-Speller), and the resulting images from the web search are shown to the user as a Rapid Serial Visual Presentations (RSVP). For each image shown, the EEG response is used by the system to recognize the user's interests and narrow down the search results. In addition, it also adds more descriptive terms to the search query, and retrieves more specific image search results and repeats the process. As a proof of concept, a prototype system was designed to test the navigation through the interface and the Hex-o-Speller using an event-related potential(ERP) detection and classification system. The results and challenges faced were noted and analyzed.

Herein a method for object tracking and interception for applications in arm robotics is presented. Generality of five and six degree of freedom robots has traditionally been lost due to focus on pre-programmed dead-reckoning centric algorithms. The method presented herein utilizes high speed software engineering techniques in conjunction with image processing algorithms to produce a viable and extensible object tracking platform.

Controlling a mobile robot using human biopotential signals has been a common problem in the field of assistive robotics. Not only it is enough to analyze the biosignal characteristics and interpret motion commands from the raw signal, but also an efficient learning algorithm may help to overcome varying characteristics of the biosignal for the sake of robust control of the mobile robot. In this work, an efficient learning algorithm utilizing Radial Basis Function Networks have been studied and applied to EOG signals in order to control a mobile robot. Obtained results show that RBF network is successful in learning the biosignal characteristics and producing sufficient control signals to control a mobile robot.

This paper proposes a hybrid algorithm that combines the principles of a traditional genetic algorithm enhanced with genetic screening and the exploration process of a scatter search algorithm to produce a novel approach. After the algorithm is presented, the algorithm is tested on a classic image compression problem using vector quantization. The goal is to show that this hybrid algorithm can be applied to a large class of problems such as clustering, function approximation, and optimization.

This paper discusses the application of particle swarm optimization (PSO) to data clustering. Four different methods of PSO are tested on six test data sets and compared to k-means and fuzzy c-means. The four PSO methods, combinations of the constriction method, inertia, and the predator-prey method all out-perform k-means and fuzzy c-means in all test cases to varying degrees in terms of quantization error.