1,720,999 research outputs found
2nd International Conference on Robot Intelligence Technology and Applications
No full textWe are facing a new technological challenge on how to store and retrieve knowledge and manipulate intelligence for autonomous services by intelligent systems which should be capable of carrying out real world tasks autonomously. To address this issue, robot researchers have been developing intelligence technology (InT) for “robots that think” which is in the focus of this book. The book covers all aspects of intelligence from perception at sensor level and reasoning at cognitive level to behavior planning at execution level for each low level segment of the machine. It also presents the technologies for cognitive reasoning, social interaction with humans, behavior generation, ability to cooperate with other robots, ambience awareness, and an artificial genome that can be passed on to other robots. These technologies are to materialize cognitive intelligence, social intelligence, behavioral intelligence, collective intelligence, ambient intelligence and genetic intelligence. The book aims at serving researchers and practitioners with a timely dissemination of the recent progress on robot intelligence technology and its applications, based on a collection of papers presented at the at the 2nd International Conference on Robot Intelligence Technology and Applications (RiTA), held in Denver, USA, December 18-20, 2013
Hybrid Cloud Model Checking Using the Interaction Layer of Harms for Ambient Assistive Living Environments
No full textSoon, humans will be co-living and taking advantage of the help of multi-agent systems in a broader way than the present. Such systems will involve machines or devices of any variety, including robots. These kind of solutions will adapt to the special needs of each individual. However, to the concern of this research effort, systems like the ones mentioned above might encounter situations that will not be seen before execution time. It is understood that there are two possible outcomes that could materialize; either keep working without corrective measures, which could lead to an entirely different end or completely stop working. Both results should be avoided, specially in cases where the end user will depend on a high level guidance provided by the system, such as in ambient intelligence applications. This dissertation worked towards two specific goals. First, to assure that the system will always work, independently of which of the agents performs the different tasks needed to accomplish a bigger objective. Second, to provide initial steps towards autonomous survivable systems which can change their future actions in order to achieve the original final goals. Therefore, the use of the third layer of the HARMS model was proposed to insure the indistinguishability of the actors accomplishing each task and sub-task without regard of the intrinsic complexity of the activity. Additionally, a framework was proposed using model checking methodology during run-time for providing possible solutions to issues encountered in execution time, as a part of the survivability feature of the systems final goals
Smart Security System Based on Edge Computing and Face Recognition
No full textPhysical security is one of the most basic human needs. People care about it for various reasons; for the safety and security of personnel, to protect private assets, to prevent crime, and so forth. With the recent proliferation of AI, various smart physical security systems are getting introduced to the world. Many researchers and engineers are working on developing AI-driven physical security systems that have the capability to identify potential security threats by monitoring and analyzing data collected from various sensors. One of the most popular ways to detect unauthorized entrance to restricted space is using face recognition. With a collected stream of images and a proper algorithm, security systems can recognize faces detected from the image and send an alert when unauthorized faces are recognized. In recent years, there has been active research and development on neural networks for face recognition, e.g. FaceNet is one of the advanced algorithms. However, not much work has been done to showcase what kind of end-to-end system architecture is effective for running heavy-weight computational loads such as neural network inferences. Thus, this study explores different hardware options that can be used in security systems powered by a state-of-the-art face recognition algorithm and proposes that an edge computing based approach can significantly reduce the overall system latency and enhance the system reactiveness. To analyze the pros and cons of the proposed system, this study presents two different end-to-end system architectures. The first system is an edge computing-based system that operates most of the computational tasks at the edge node of the system, and the other is a traditional application server-based system that performs core computational tasks at the application server. Both systems adopt domain-specific hardware, Tensor Processing Units, to accelerate neural network inference. This paper walks through the implementation details of each system and explores its effectiveness. It provides a performance analysis of each system with regard to accuracy and latency and outlines the pros and cons of each system
A Monocular Vision-Based Target Surveillance and Interception System Demonstrated in a Counter Unmanned Aerial System (CUAS) Application
No full textSeveral events in the recent years involving hobby Unmanned Aerial Vehicles (UAVs), colloquially known as drones, have demonstrated a general lack in security regarding relatively slow moving, low flying aerial objects. Additionally, these events have demonstrated the ease of using hobby UAVs for illegal or illicit purposes. Although researchers and industry professionals are working simultaneously to find solutions to detecting hobby UAVs and countering any possible threats, there are no options available, at the time of this study, that provide a practical solution to the UAV problem. This dissertation presents a possible solution to the Counter Unmanned Aerial System (CUAS) problem. This dissertation presents an autonomous system, developed within the framework of the Robotic Operating System (ROS), that uses only a monocular camera on board a UAV to autonomously detect, track, and follow an unauthorized UAV for surveillance purposes. If the UAV is determined to be a threat, the UAV, under the control of the autonomous system, uses itself as a counter defense by targeting and intercepting the flight of the unauthorized UAV, causing an air-to-air collision that results in a kinetic kill of the unauthorized UAV. The surveillance mode of the autonomous system described within this dissertation was evaluated on the capability to follow the unauthorized UAV for a distance that is greater than 50% of the total course distance. The interception mode of the system was evaluated on the number of direct interceptions of the unauthorized UAV. Results of the experiments conducted to test both modes of the autonomous system showed that when the unauthorized UAV was detected, the surveillance mode of the autonomous system typically tracked and followed the target for an average 87.23% of the average total distance of the courses in the experiments. Additionally, the experiments resulted in a mere 6.7% of unintentionally interceptions during the surveillance experiments, which demonstrates the control of the system despite a lack of distance information available. Results for the interception mode showed that the system presented here had a direct interception rate of 40%. If a UAV were carrying a net, the rate of interception increases to 81.3%
EDGE Computing Approach to Indoor Temperature Prediction Using Machine Learning
No full textThis paper aims to present a novel approach to real-time indoor temperature forecasting to meet energy consumption constraints in buildings, utilizing computing resources available at the edge of a network, close to data sources. This work was inspired by the irreversible effects of global warming accelerated by greenhouse gas emissions from burning fossil fuels. As much as human activities have heavy impacts on global energy use, it is of utmost importance to reduce the amount of energy consumed in every possible scenario where humans are involved. According to the US Environmental Protection Agency (EPA), one of the biggest greenhouse gas sources is commercial and residential buildings, which took up 13 percent of 2019 greenhouse gas emissions in the United States. In this context, it is assumed that information of the building environment such as indoor temperature and indoor humidity, and predictions based on the information can contribute to more accurate and efficient regulation of indoor heating and cooling systems. When it comes to indoor temperature, distributed IoT devices in buildings can enable more accurate temperature forecasting and eventually help to build administrators in regulating the temperature in an energy-efficient way, but without damaging the indoor environment quality. While the IoT technology shows potential as a complement to HVAC control systems, the majority of existing IoT systems integrate a remote cloud to transfer and process all data from IoT sensors. Instead, the proposed IoT system incorporates the concept of edge computing by utilizing small computer power in close proximity to sensors where the data are generated, to overcome problems of the traditional cloud-centric IoT architecture. In addition, as the microcontroller at the edge supports computing, the machine learning-based prediction of indoor temperature is performed on the microcomputer and transferred to the cloud for further processing. The machine learning algorithm used for prediction, ANN (Artificial Neural Network) is evaluated based on error metrics and compared with simple prediction models
Cost Optimization of the Unmanned Aircraft Delivery System with Public Transportation
No full textDrone delivery systems have been popular in recent days to reduce the cost of delivery, and many studies showed that the delivery cost is also a critical factor the customers consider. However, drone delivery systems have many limitations to commercialize because UAVs(Unmanned Aerial Vehicles) have limited capacity, limited battery life, safety issues and government regulations. This paper focuses on saving the battery life of UAVs to make drone delivery systems feasible. The battery life of UAVs is hard to check since the battery is affected by numerous factors like weather, the weight of packages and velocity. This paper minimizes the total distance of UAVs travel which is one of the major energy factors to save energy. If a UAV can cooperate with existing public transportation, systems can save battery life and can deliver multiple packages simultaneously. On top of that, if the public transportation has charging panels, a UAV can even charge its battery while it takes the public transportation. This paper addresses the problem of \say{What if UAVs cooperate with the current bus systems? Given the destinations, which bus will UAV take and how to decide the destination to visit first? How many UAVs are needed to deliver the packages?}, By running a simulation with real bus data from CityBus and randomly generated destinations. This paper suggests several algorithms save the total cost of the delivery and shows two experiments to answer above questions. The first experiment shows that the effectiveness of the UAVs and bus co-operating by comparing the total distance of UAVs. The treatment group composed of UAVs and bus shares the destinations then optimizes the route(the order of the destinations to visit). The second experiment assumes that the drone delivery systems already have the optimal route and calculates the number of UAVs needed for the given routes. The result of the experiments shows that the drone delivery systems can save the total distance of the UAVs travel by cooperating with the existing public transportation systems. Then, the constraint of battery life and energy is reduced
Low-Cost and Scalable Visual Drone Detection System Based on Distributed Convolutional Neural Network
No full textRecently, with the advancement in drone technology, more and more hobby drones are being manufactured and sold across the world. However, these drones can be repurposedfor the use in illicit activities such as hostile-load delivery. At the moment there are not many systems readily available for detecting and intercepting those hostile drones. Although there is a prototype of a working drone interceptor system built by the researchers of Purdue University, the system was not ready for the general public due to its nature of proof-of-concept and the high price range of the military-grade RADAR used in the prototype. It is essential to substitute such high-cost elements with low-cost ones, to make such drone interception system affordable enough for large-scale deployment.This study aims to provide an alternative, affordable way to substitute an expensive, high-precision RADAR system with Convolutional Neural Network based drone detection system, which can be built using multiple low-cost single board computers. The experiment will try to find the feasibility of the proposed system and will evaluate the accuracy of the drone detection in a controlled environment.</div
DISASTER RELIEF SUPPLY MODEL FOR LOGISTIC SURVIVABILITY
No full textDisasters especially from natural phenomena are inevitable. The affected areas recover from the aftermath of a natural disaster with the support from various agents participating in humanitarian operations. There are several domains of the operation, and distributing relief aids is one. For distribution, satisfying the demand for relief aid is important since the condition of the environment is unfavorable to affected people and resources needed for the victim’s life are scarce. However, it becomes problematic when the logistic agents believed to be work properly fail to deliver the emergency goods because of the capacity loss induced from the environment after disasters. This study was proposed to address the problem of logistic agents’ unexpected incapacity which hinders scheduled distribution. The decrease in a logistic agent’s supply capability delaysachieving the goal of supplying required relief goods to the affected people which further endangers them. Regarding the stated problem, this study explored the importance ofsetting the profile of logistic agents that can survive for certain duration of times. Therefore, this research defines the “survivability” and the profile of logistic agents for surviving the last mile distribution through agent based modeling and simulation. Through simulations, this study uncovered that the logistic exercise could gain survivability with the certain number and organization of logistic agents. Proper formation of organization establish the logistics’ survivability, but excessive size can threaten the survivability
Sentiment Analysis Using Bert on Yelp Restaurant Reviews
No full textYelp is a platform for users to leave text-based reviews of products or services in addition to photos and ratings from one to five stars. This study addresses two distinct problems that Yelp currently has. First, Yelp\u27s exorbitant number of text-based reviews sometimes makes it impossible for the user to go through and read every single review. Second, the lack of specificity of Yelp\u27s current one-to-five-star rating system cannot determine the rationales of the customers if they have given the same rating. To solve the aforementioned problems, the study focused on the initial stage of the algorithm by answering the research question, Can the BERT model determine whether a customer\u27s review on Yelp is positive or negative, and the degree of said positivity or negativity, based on the review\u27s content? . To answer the stated research question, the study provided each step of the research approach: (1) tokenization and removing stop words, (2) keyword analysis, (3) preparation for the BERT model, and (4) training the BERT model. Based on the results obtained from the research approach, the study supported the research question that the researcher established in this study. The researcher concluded the study by summarizing the limitations of the study and introducing the future development algorithm that would be focused on building on this initial stage to assign a ranking on a one-to-five scale of each pre-defined category based on the contents of the text-based reviews
Gas Source Declaration Using a Mobile Robot and a Machine Learning Algorithm in Indoor Environments
No full textThis paper introduces a novel algorithm for gas source declaration, which is the third task of the gas source localization problem. With the algorithm, a robot should be able to determine whether the robot is located near the gas source or not. However, due to the turbulence in the gas distribution in indoor environments, lacking a strong airflow, no analytical model can be applied to the gas distribution, and it is tough to analyze the characteristics of the gas distribution. Therefore, to identify and classify the characteristics of the gas distribution, nonlinear classification algorithms are employed. Also, four feature extraction methods will be used to generate features from the gas distribution. Based on those features, the classification performance and the speed of five classifiers (MLP, KNN, SVM, AdaBoost, and Voting) were evaluated. Also, the most appropriate parameters of each classifier under the each testing environment were suggested. Using those features and parameters, the given classifiers showed the maximum success rates approximately 90.8 \% and 85.3 \% in the large and small testing rooms, respectively
- …
