{"title":"Accurate Gridless Indoor Localization Based on Multiple Bluetooth Beacons and Machine Learning","authors":"Konstantinos Kotrotsios, T. Orphanoudakis","doi":"10.1109/ICARA51699.2021.9376476","DOIUrl":"https://doi.org/10.1109/ICARA51699.2021.9376476","url":null,"abstract":"In this work we present an indoor location method using smartphones as a source of location information. The proposed method uses the Received Signal Strength Indicator (RSSI) value from Bluetooth Low Energy Beacons scattered around interior spaces. We present the results of our model using machine learning, which was developed based on measurements of RSSI values from Beacons inside a lab environment occupying a space of 31m2. Measurements were fed to the open-source TensorFlow framework to develop an estimator of the distance between the mobile phone and the beacon. Next, based on the cross-sections of peripheral lines having as a center the location of the Beacons and radius the predicted distances we compute the intersection points from all circles and base our position estimation on the Geometric median of intersection points. Through experiments, we show that our system has an average accuracy of 69.58cm and can predict position with an accuracy of less than a meter in 80% of the cases.","PeriodicalId":183788,"journal":{"name":"2021 7th International Conference on Automation, Robotics and Applications (ICARA)","volume":"90 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123458537","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Joyraj Bhowmick, Anurag Singh, Harshit Gupta, R. Nallanthighal
{"title":"A Novel Approach to Computationally Lighter GNSS-Denied UAV Navigation Using Monocular Camera","authors":"Joyraj Bhowmick, Anurag Singh, Harshit Gupta, R. Nallanthighal","doi":"10.1109/ICARA51699.2021.9376502","DOIUrl":"https://doi.org/10.1109/ICARA51699.2021.9376502","url":null,"abstract":"A good receptivity of GNSS signals during autonomous navigation is of prime importance. Due to this, navigation becomes close to impossible with places having low receptivity or GNSS-denied zones. In this paper, a navigational system is presented for a multi-rotor UAV in a GNSS-denied environment. The global positioning problem is solved by taking a local reference frame and estimating the UAV's position and velocity w.r.t it, using an onboard monocular camera and barometer. The navigational problem is solved by designing a required velocity function and tuning the system's response to this input by a PID controller. The results of both simulations and actual flight tests show the effectiveness and reliability of the presented work, which enables a UAV to perform full automated missions from takeoff, waypoint navigation and landing, in complete absence of any GNSS signals.","PeriodicalId":183788,"journal":{"name":"2021 7th International Conference on Automation, Robotics and Applications (ICARA)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125526884","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Dynamic Cognitive-Social Particle Swarm Optimization","authors":"Khelil Kassoul, S. Belhaouari, N. Cheikhrouhou","doi":"10.1109/ICARA51699.2021.9376550","DOIUrl":"https://doi.org/10.1109/ICARA51699.2021.9376550","url":null,"abstract":"Particle Swarm Optimization (PSO) is a heuristic optimization algorithm based on the modeling of the behavior of fishes and birds flock. This paper proposes a better version of PSO, named Dynamic Cognitive-Social PSO “DCS-PSO”, for global minima search by introducing optimal and dynamic cognitive and social scaling parameters without taking into consideration the inertia term. Furthermore, the velocity of each particle is controlled systematically at each iteration to avoid local minimum traps and to converge quickly and reliably towards the global minimum. The proposed algorithm is more suitable for high dimensional optimization problems and it has gotten over the limitations of classical Particle Swarm Optimization. Several experiments have been carried out, using the proposed DCS-PSO, to optimize thirteen benchmark functions and an important improvement has been observed, not only in terms of reaching the best global solutions but also in terms of convergence speed, compared to the existing benchmark approaches.","PeriodicalId":183788,"journal":{"name":"2021 7th International Conference on Automation, Robotics and Applications (ICARA)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122335451","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
I. Guarneri, G. Messina, A. Bruna, Davide Giacalone
{"title":"A Deep Learning Short Commands Recognition for MCU in Robotics Applications","authors":"I. Guarneri, G. Messina, A. Bruna, Davide Giacalone","doi":"10.1109/ICARA51699.2021.9376366","DOIUrl":"https://doi.org/10.1109/ICARA51699.2021.9376366","url":null,"abstract":"This paper presents an application for vocal commands recognition based on neural network and running on a Microcontroller Unit (MCU). The recognized commands are sent via Bluetooth Low Energy to a robot which executes the command to perform movements. The aim of this work is to enable the deep learning for embedded devices presenting a neural network model running on a microcontroller with limited memory and clock frequency.","PeriodicalId":183788,"journal":{"name":"2021 7th International Conference on Automation, Robotics and Applications (ICARA)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122086035","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Towards a Highly Integrated 3D Printed Soft Continuum Manipulator","authors":"J. M. Salgueiro, J. Reis","doi":"10.1109/ICARA51699.2021.9376438","DOIUrl":"https://doi.org/10.1109/ICARA51699.2021.9376438","url":null,"abstract":"Soft continuum robot manipulators are structurally compliant devices that aim to take advantage of this characteristic to facilitate interaction with the surrounding environment. However there is a delicate balance between structural compliance and the ability to perform useful mechanical work. Off-axis manipulation has been identified in literature as one of the challenges posed to these robots. In this paper the design and experimental evaluation of a soft continuum manipulator robot module, based on a soft wave spring structure, is addressed with the aim of studying and improving off-axis manipulation. Finite element simulation results are presented, that allow to optimize the design of the wave spring for a specific set of dimensional constraints. A working prototype is also described, and experimental results are presented regarding the evaluation of the structure in both in-axis and off-axis loading, and also for trajectory following under several load cases.","PeriodicalId":183788,"journal":{"name":"2021 7th International Conference on Automation, Robotics and Applications (ICARA)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121597155","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Olimpiya Saha, Guohua Ren, Javad Heydari, Viswanath Ganapathy, Mohak Shah
{"title":"Deep Reinforcement Learning Based Online Area Covering Autonomous Robot","authors":"Olimpiya Saha, Guohua Ren, Javad Heydari, Viswanath Ganapathy, Mohak Shah","doi":"10.1109/ICARA51699.2021.9376477","DOIUrl":"https://doi.org/10.1109/ICARA51699.2021.9376477","url":null,"abstract":"Autonomous area covering robots have been increasingly adopted in for diverse applications. In this paper, we investigate the effectiveness of deep reinforcement learning (RL) algorithms for online area coverage while minimizing the overlap. Through simulation experiments in grid based environments and in the Gazebo simulator, we show that Deep Q-Network (DQN) based algorithms efficiently cover unknown indoor environments. Furthermore, through empirical evaluations and theoretical analysis, we demonstrate that DQN with prioritized experience replay (DQN-PER) significantly minimizes the sample complexity while achieving reduced overlap when compared with other DQN variants. In addition, through simulations we demonstrate the performance advantage of DQN-PER over the state-of-the-art area coverage algorithms, BA* and BSA. Our experiments also indicate that a pre-trained RL agent can efficiently cover new unseen environments with minimal additional sample complexity. Finally, we propose a novel way of formulating the state representation to arrive at an area-agnostic RL agent for efficiently covering unknown environments.","PeriodicalId":183788,"journal":{"name":"2021 7th International Conference on Automation, Robotics and Applications (ICARA)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126246210","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"An RRT-Based Path Planning Strategy in a Dynamic Environment","authors":"Yijing Li","doi":"10.1109/ICARA51699.2021.9376472","DOIUrl":"https://doi.org/10.1109/ICARA51699.2021.9376472","url":null,"abstract":"The real-time RRT-based path planning strategy is designed for the non-holonomic robot in a dynamic environment. The sampling-based strategy, which consists of a pre-processing RRT path planner and a real-time planner, navigate the robot to avoid the unknown moving obstacle, which is time-varying or move randomly. Additionally, the algorithm contains a simple temporary target determination function, and ensures its feasibility in the target-unknown situation. It decreases the realtime computational complexity because of the omission of moving obstacle segmentation, velocity computation, or original path replanning. The feasibility of the navigation strategy is verified by using computation simulation.","PeriodicalId":183788,"journal":{"name":"2021 7th International Conference on Automation, Robotics and Applications (ICARA)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130187448","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Qizhang Lin, Yan Ding, Hong Xu, Wenxiang Lin, Jiaxin Li, Xiaoxiao Xie
{"title":"ECascade-RCNN: Enhanced Cascade RCNN for Multi-scale Object Detection in UAV Images","authors":"Qizhang Lin, Yan Ding, Hong Xu, Wenxiang Lin, Jiaxin Li, Xiaoxiao Xie","doi":"10.1109/ICARA51699.2021.9376456","DOIUrl":"https://doi.org/10.1109/ICARA51699.2021.9376456","url":null,"abstract":"Due to the change of flight altitude and attitude of UAV, the object scale in UAV images exists difference which leads to a great challenge for object detection and has drawn wide attention. In this paper, an improved object detection network named ECascade-RCNN is proposed to deal with the multi-scale problem in object detection task for UAV images. We present an innovative Trident-FPN backbone to extract features and design a new attention mechanism to enhance the performance of the detector. Moreover, k-means algorithm is adapted to generate anchors so that the detection model can get better regression accuracy. We evaluate the proposed ECascade-R-CNN on Visdrone dataset through several ablation experiments and the results show that the ECascade-RCNN given in the paper is effective. The ECascade-RCNN is also used in the Visdrone2020 challenge and ranked 8th on the object detection track.","PeriodicalId":183788,"journal":{"name":"2021 7th International Conference on Automation, Robotics and Applications (ICARA)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128843869","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Self-Repairing Line of Metamorphic Robots","authors":"Nooshin Nokhanji, N. Santoro","doi":"10.1109/ICARA51699.2021.9376447","DOIUrl":"https://doi.org/10.1109/ICARA51699.2021.9376447","url":null,"abstract":"A Metamorphic Robots System is a modular self-reconfigurable robotic system composed of autonomous mobile modules in a 2D (or 3D) regular grid. The modules have limited computational capabilities, interact only with neighboring modules, and can move around adjacent modules from a cell to an empty neighboring cell under specific conditions. An important well-studied problem for these robotic systems is Motion Planning, also known as Shape Formationor Self-reconfiguration, requiring the modules to organize themselves into a pre-determined final configuration (i.e., shape); basic shapes such as the line (or chain) are especially important as they are utilized as a foundation for constructing more complicated shapes and are an initial measure for handling complicated tasks. A metamorphic robots system could offer a higher degree of reliability and robustness compared to fixed-architecture robots due to its capacity to self-repair: should some modules fail and no longer move, the shape could be reconstructed by the non-faulty modules. To do this correctly, efficiently, and without restricting the autonomy of the modules is a non-trivial task. In this paper, we study the Line Recoveryproblem, requiring the non-faulty modules to reconstruct the line without violation of connectivity requirements at any time during the recovery procedure. A thorough feasibility characterization of the problem, the necessary conditions for its solvability, and an algorithm that solves the problem, regardless of the number and distribution of faults, are provided.","PeriodicalId":183788,"journal":{"name":"2021 7th International Conference on Automation, Robotics and Applications (ICARA)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127806991","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Hardware in the Loop Simulation of Aircraft Inspection by an Unmanned Aerial System","authors":"Daniel Dose, M. Tappe, M. Alpen, J. Horn","doi":"10.1109/ICARA51699.2021.9376532","DOIUrl":"https://doi.org/10.1109/ICARA51699.2021.9376532","url":null,"abstract":"Inspection of commercial aircrafts, wind turbines, bridges and other infrastructure elements is done manually in many cases. Therefore, today's maintenance mostly is time-consuming and cost-intensive. The goal of the joint project AI inspection drone is the holistic system design of an unmanned aerial system (UAS) for the damage detection and assessment of airliners. The technical design should be based on current maintenance requirements and the overall system should be able to anticipate its own flight, evaluate the inspection data AI-based, and draw automatic conclusions. In this paper we describe the structure of the announced and partly already realized process chain. The focus is on the required interfaces between the individual components, the control engineering challenges to the UAS and the mapping of the entire process chain in a simulation environment, which also enables a hardware in the loop test of the different sensors and the carrier system itself. Based on this simulation which also includes the mapping of the technical and operational environment, different movement strategies with regard to energy requirements and flight time as well as an efficient sensor data fusion should be investigated. The results obtained are, as far as possible, validated by simulation and real experiments.","PeriodicalId":183788,"journal":{"name":"2021 7th International Conference on Automation, Robotics and Applications (ICARA)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114966252","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}