In this paper, a new multi-robot Adaptive Search Space Coverage Path Planning (multi-robot ASSCPP) approach is proposed to explore and cover 3D large structures using a decentralized multi-robot system. The multi-robot ASSCPP approach divides the coverage load on a team of robots to decrease the CPP execution time and simultaneously achieve high coverage. It utilizes the existence of the reference mesh model to divide it into segments based on surface area. Then it generates viewpoints adaptively exploiting the sensor's noise models to direct the search towards areas with low resolution and low coverage. The coverage paths are then generated using a heuristic function that evaluates the traveling distance and the quality of the model. In the multi-robot system, the generated coverage paths are executed on each robot starting at their allocated path starting position. Experiments were conducted in a realistic robotic simulator to test the validity of the proposed algorithm.
Robustly tracking a person of interest in the crowd with a robotic platform is one of the cornerstones of human-robot interaction. The robot platform which is limited by the computational power, rapid movements and occlusions of target requires an efficient and robust framework to perform tracking. This paper proposes a deep learning framework for tracking a person using a mobile robot with stereo camera. The proposed system detects a person based on its head, then utilizes the low cost, high speed regression network based tracker to track the person of interest in real time. The visual servoing of the mobile robot has been designed using PID controller which utilizes tracker output and depth estimation of the person in subsequent frames, hence providing smooth and adaptive movement of the robot based on target movement. The proposed system has been tested in real environment, thus proving its effectiveness.
In this paper, we propose a robotic gripper that utilizes a soft finger based on the Fin Ray Effect and a neuromorphic vision-based sensor to obtain tactile information from the finger. The sensor captures tactile information by observing brightness intensity changes of markers that are placed internally inside the finger. The tactile information is processed to detect incipient slip at a rate of 500?s.
The analysis, implementation and control of modular multi-copter structures, capable of coordinated flight, are investigated in this work. Graph theory is employed in order to generically define arbitrary multi-copter assemblies and extract dynamic properties. The proposed controller works by viewing each copter as an agent capable of providing a total thrust force and yaw torque, and can be implemented on most commercial flight control units, using standardized methods. Experimental studies are offered to show positive initial results, while future research goals are discussed, taking into account structural flexibilities and moving towards flexible aerial robots.
Current state-of-the-art indoor positioning involves agent-carrying Global Positioning System (GPS) service. However, such system does not provide the needed sufficient accuracy. On the other hand, the Dynamic Vision Sensor (DVS) is becoming an attraction in the robotic perception field. This paper proposes a Visual Light Positioning (VLP) system architecture facilitated by a DVS and flickering LEDs at high frequencies for stationary and in-motion indoor Unmanned Vehicles (UVs). Besides, a pixel-level frequency-based filtering algorithm is developed for the given application. The high dynamic range and low latency of the DVS allows it to detect high frequency blinking and grants it high potential to be utilized in VLP systems. Since the DVS requires brightness change in the scene, flickering LEDs overcome this obstacle and allow continuous event generation even when the robotic agent is stationary.
Robotic vision has a major role in robotic manipulation in industry to assist and serve different applications. The traditional frame-based visual servoing has several limitations in visual feedback for robotic manipulation due to their operation method of capturing scenes continuously even if no changes in the scene is occurring. Alternatively, event cameras detects dynamic changes asynchronously at a high temporal resolution (1μs) with low latency and wide dynamic range hence, it is more of interest. In this project, I present visual sevoing method using event camera to achieve manipulation task. Additionally, to compensate for the inaccuracies and limitations of visual servoing, an event-driven tactile sensor is capable of fine manipulation with low-latency, less computational cost and power consumption. The event based visual servoing (EBVS) is validated experimentally using a commercial robot manipulator with an eye-in-hand configuration. The experiments proves the success of the EBVS method with a 100% success rate.
The onset of the COVID-19 pandemic necessitated wearing personal protective equipment (PPE) to prevent the spread of the SARS-CoV-2 virus. Amongst them, the wearing of face masks was recognized as one of the most crucial ways to reduce infection, as the mask works as filter for the different microorganisms. In this work the geometrical part of the filtration process of the N95 and surgical masks was studied using luminescent ultra-small silicon nanoparticles (Si-NPs) to represent the SARS-CoV-2 by spraying it on the mask using an atomizer. Scanning electron microscopy (SEM), and optical microscope were used to check the masks after depositing the Si-NPs. The obtained images show that the Si nanoparticles are trapped by the PE fiber network, indicating its ability to filter the SARS-CoV-2. This visualization using nanotechnology can help to further improve mask designs for better filtration.
Graphene oxide is one of the most interesting materials in research today due to its superior properties. In this study, different concentrations of graphene oxide (GO) were used to improve the hydrophilicity of the COC surface, thereby enhancing the wettability of its patterned microchannels. Argon-oxygen plasma was used to improve the adhesion between GO and COC surface. Plasma power of 30W with exposure times from 5min to 2hrs were carried out.
This work presents the exploration of the rare earth high entropy oxides based catalysts: 5%Ni-500/CeGdLaPrSmO-CP, 10%Ni-500/CeGdLaPrSmO-CP, 10%Ni-900/CeGdLaPrSmO-CP and 15%Ni-500/CeGdLaPrSmO-CP and shades the light on investigating their catalytic activity for CO2 hydrogenation to methane reaction and explains their performance based on their morphological and textural properties. The high entropy oxide support CeGdLaPrSmO that crystalized into fluorite structure was synthesized by coprecipitation method and the nickel was added to it by the wet impregnation. The nickel loaded high entropy oxide based catalysts exhibited maximum activity that reached up to 51.2 % of converting CO2 at 500 °C.
This study presents a one-dimensional model for nanofluid flow and heat transfer with nanoparticle aggregation. The model is governed by mass, momentum, energy and nanoparticle species conservation. The model is applied to investigate (1) nanoparticle aggregation under a quiescent condition and (2) heat transfer performance of nanofluid flow in pipes.
