Search

This paper presents a Machine Learning approach for the classification of Amazonian fruits (Moriche, Asai and Seje). Vegetative indices were used as features to drive the corresponding classification by processing RGB/VIS imagery. In this regard, we used four Machine Learning models to identify the stage of maturity for the fruits: Multi-variable regressions, Naives Bayes, Support Vector Machines and Artificial Neural Networks. These models were trained and tested with the features of each variety. Experimental results were validated by calculating ROC data, in which neural networks achieved an accuracy of 99% in the stage of maturity identification for the three amazonian varieties. These results allow us to conclude that the used vegetative indices accurately correlate with the physiological characteristics of the fruits, being relevant for the stage of maturity of the three varieties.

This paper proposes an integrated method for the estimation of soil moisture in potato crops that uses a low-cost wireless sensor network (WSN). Soil moisture estimation maps were created by applying the Kriging technique over a WSN composed of 11×11 nodes. Our goal is to estimate the soil moisture of the crop with a small-scale WSN. Using a perfect mesh approach on a potato crop, experimental results demonstrated that 25 WSN nodes were optimal and sufficient for soil moisture characterization, achieving estimations errors

High-throughput platforms for plant phenotyping usually demand expensive high-density LiDAR devices with computational intense methods for characterizing several morphological variables. In fact, most platforms require offline processing to achieve a comprehensive plant architecture model. In this paper, we propose a low-cost plant phenotyping system based on the sensory fusion of low-density LiDAR data with multispectral imagery. Our contribution is twofold: (i) an integrated phenotyping platform with embedded processing methods capable of providing real-time morphological data, and (ii) a multi-sensor fusion algorithm that precisely match the 3D LiDAR point-cloud data with the corresponding multispectral information, aiming for the consolidation of four-dimensional plant models. We conducted extensive experimental tests over two plants with different morphological structures, demonstrating the potential of the proposed solution for enabling real-time plant architecture modeling in the field, based on low-density LiDARs.

Traditional methods to measure spatio-temporal variations in above-ground biomass dynamics (AGBD) predominantly rely on the extraction of several vegetation-index features highly associated with AGBD variations through the phenological crop cycle. This work presents a comprehensive comparison between two different approaches for feature extraction for non-destructive biomass estimation using aerial multispectral imagery. The first method is called GFKuts, an approach that optimally labels the plot canopy based on a Gaussian mixture model, a Montecarlo-based K-means, and a guided image filtering for the extraction of canopy vegetation indices associated with biomass yield. The second method is based on a Graph-Based Data Fusion (GBF) approach that does not depend on calculating vegetation-index image reflectances. Both methods are experimentally tested and compared through rice growth stages: vegetative, reproductive, and ripening. Biomass estimation correlations are calculated and compared against an assembled ground-truth biomass measurements taken by destructive sampling. The proposed GBF-Sm-Bs approach outperformed competing methods by obtaining biomass estimation correlation of 0.995 with R2=0.991 and RMSE=45.358 g. This result increases the precision in the biomass estimation by around 62.43% compared to previous works.

As technology and legislation move forward (JAA & Eurocontrol, 2004) remotely controlled, semi-autonomous or autonomous Unmanned Aerial Systems (UAS) will play a significant role in providing services and enhancing safety and security of the military and civilian community at large (e.g. surveillance and monitoring) (Coifman et al., 2004). The potential market for UAVs is, however, much bigger than just surveillance. UAVs are ideal for risk assessment and neutralization in dangerous areas such as war zones and regions stricken by disaster, including volcanic eruptions, wildfires, floods, and even terrorist acts. As they become more autonomous, UAVs will take on additional roles, such as air-to-air combat and even planetary science exploration (Held et al., 2005). As the operational capabilities of UAVs are developed there is a perceived need for a significant increase in their level of autonomy, performance, reliability and integration with a controlled airspace full of manned vehicles (military and civilian). As a consequence researchers working with advanced UAVs have moved their focus from system modeling and low-level control to mission planning, supervision and collision avoidance, going from vehicle constraints to mission constraints (Barrientos et al., 2006). This mission-based approach is most useful for commercial applications where the vehicle must accomplish tasks with a high level of performance and maneuverability. These tasks require flexible and powerful trajectory-generation and guidance capabilities, features lacking in many of the current commercial UAS. For this reason, the purpose of this work is to extend the capabilities of commercially available autopilots for UAVs. Civil systems typically use basic trajectory-generation algorithms, capable only of linear waypoint navigation (Rysdyk, 2003), with a minimum or non-existent control over the trajectory. These systems are highly constrained when maneuverability is a mission requirement. On the other hand, military researchers have developed algorithms for high-performance 3D path planning and obstacle avoidance (Price, 2006), but these are highly proprietary technologies that operate with different mission constraints (target acquisition, threat avoidance and situational awareness) so they cannot be used in civil scenarios.

Monitoring large-scale crops using an autonomous UAV demands optimal path planning methods to increase remote sensing autonomy. This paper presents a comprehensive optimization model to minimize the UAV battery consumption during crop monitoring tasks.

Image segmentation enables the precise extraction of several crop traits from multispectral aerial imagery. This paper presents a novel segmentation technique called GFKuts. The method integrates a graph-based optimization algorithm with a k-means Monte Carlo approach. Here, we evaluate the performance of the proposed method against other approaches for image segmentation found in the specialized literature. Results report an improvement on the F1-score accuracy in terms of crop canopy segmentation. These findings are promising for the precise calculation of vegetative indices and other crop trait features.

Plant architecture is defined as the three-dimensional modeling of the plant’s morphology for extracting relevant phenological traits. Most applications rely on expensive high-density LiDAR devices for enabling high-throughput mapping. In this paper, we explore the use of low-cost LiDAR equipment by using a sensor fusion approach. The proposed method is based on the fusion of LiDAR-acquired low resolution 3D point cloud data with high resolution 2D imagery. We use an extrinsic calibration method that requires oversampling to enhance the data fusion from both sensors. As a result, we increased the resolution of the output 3D model of the plant.

This paper reports the integration of a remotely operated vehicle (ROV) solution for monitoring water quality in fish farms. The robotic system includes a RGB camera for real-time video capturing and a set of integrated sensors to measure hydro-climatic data. Computer vision algorithms were implemented with the aim of inspecting net-cages in fish farms. A comprehensive software solution was developed to allow a seamless use of the vision algorithms proposed in this work. Our system was designed to process underwater imagery captured by the ROV in order to determine net patterns associated with net failure. The system was tested in a dam under real conditions. ROC data were computed to demonstrate the accuracy of the proposed system during underwater fish cage inspection. On average, we obtained an accuracy of 0.91 regarding net pattern reconstruction tasks, while an accuracy of 0.79 for net damage detection under different underwater scenarios.

Robotic-assisted systems have been playing a key role in improving and speeding up motor recovery during stroke rehabilitation therapies. This paper presents an approach to determine velocity patterns based on the analysis of the EMG muscular condition of the hand. To this purpose, we conducted an experimental protocol with 18 subjects participating as volunteers, with the aim of acquiring EMG signals for three levels of the muscular condition: non-fatigue, transition-to-fatigue, and fatigue. Artificial Neural Networks (ANN) were trained to identify the aforementioned muscular condition levels, while a Sugeno-Type Fuzzy Inference system was used to determine the velocity based on the output of the ANN classifiers. Results indicate the proposed approach can be used for the accurate modulation of pinch-grip therapies according to the muscular condition. These are promising results towards the development of EMG-driven robotic-assistance rehabilitation therapies for stroke patients.

Robotic rehabilitation has been proposed as a promising alternative in recovery after stroke, which still presents many challenges. We present here an initial approach to a progressive robot-assisted hand-motion therapy. Firstly, our system identifies finger motion patterns from electromyographic (EMG) signals of 20 control volunteers during 5 hand exercises commonly used in rehabilitation. Secondly, the system characterizes 3 muscular condition levels, using muscular contraction strength, co-activation level and muscular activation level measurements. We compared the performance of Artificial Neural Networks (ANN), Support Vector Machines (SVM), Linear Discriminant Analysis Classifier (LDA) and kNearest Neighbor (k-NN) algorithms to classify the 5 gestures and 3 levels. Thirdly, each identified gesture and level was mapped into a spatial trajectory of an exoskeleton model, using a generalization of joint trajectories from subjects and a posterior interpolation. The statistical analysis between 36 different classifier architectures showed that a SVM classifier (cubic kernel) had the best performance to identify the 15 classes (F-score of 0.8 on average). Furthermore, the average correlation between the generated spatial trajectories and the tracked hand-motion was 0.89. In the future, the trajectories controlled by EMG signals could drive the exoskeleton for rehabilitation patients.

International audience; Traditional methods to measure spatio-temporal variations in biomass rely on a labor-intensive destructive sampling of the crop. In this paper, we present a high-throughput phenotyping approach for the estimation of Above-Ground Biomass Dynamics (AGBD) using an unmanned aerial system. Multispectral imagery was acquired and processed by using the proposed segmentation method called GFKuts, that optimally labels the plot canopy based on a Gaussian mixture model, a Montecarlo based K-means, and a guided image filtering. Accurate plot segmentation results enabled the extraction of several canopy features associated with biomass yield. Machine learning algorithms were trained to estimate the AGBD according to the growth stages of the crop and the physiological response of two rice genotypes under lowland and upland production systems. Results report AGBD estimation correlations with an average of r = 0.95 and R 2 = 0.91 according to the experimental data. We compared our segmentation method against a traditional technique based on clustering. A comprehensive improvement of 13% in the biomass correlation was obtained thanks to the segmentation method proposed herein.

Leaf nitrogen (N) directly correlates to chlorophyll production, affecting crop growth and yield. Farmers use soil plant analysis development (SPAD) devices to calculate the amount of chlorophyll present in plants. However, monitoring large-scale crops using SPAD is prohibitively time-consuming and demanding. This paper presents an unmanned aerial vehicle (UAV) solution for estimating leaf N content in rice crops, from multispectral imagery. Our contribution is twofold: (i) a novel trajectory control strategy to reduce the angular wind-induced perturbations that affect image sampling accuracy during UAV flight, and (ii) machine learning models to estimate the canopy N via vegetation indices (VIs) obtained from the aerial imagery. This approach integrates an image processing algorithm using the GrabCut segmentation method with a guided filtering refinement process, to calculate the VIs according to the plots of interest. Three machine learning methods based on multivariable linear regressions (MLR), support vector machines (SVM), and neural networks (NN), were applied and compared through the entire phonological cycle of the crop: vegetative (V), reproductive (R), and ripening (Ri). Correlations were obtained by comparing our methods against an assembled ground-truth of SPAD measurements. The higher N correlations were achieved with NN: 0.98 (V), 0.94 (R), and 0.89 (Ri). We claim that the proposed UAV stabilization control algorithm significantly improves on the N-to-SPAD correlations by minimizing wind perturbations in real-time and reducing the need for offline image corrections.

This paper presents an integrated aerial system for the identification of Amazonian Moriche palm (Mauritia flexuosa) in dense forests, by analyzing the UAV-captured RGB imagery using a Mask R-CNN deep learning approach. The model was trained with 478 labeled palms, using the transfer learning technique based on the well-known MS COCO framework©. Comprehensive in-field experiments were conducted in dense forests, yielding a precision identification of 98%. The proposed model is fully automatic and suitable for the identification and inventory of this species above 60 m, under complex climate and soil conditions.

This paper presents an approach for explosive-landmine detection by designing, developing and integrating a Ground Penetrating Radar (GPR) onboard an autonomous aerial drone. Our goal is twofold: f...

This paper addresses the Nonlinear Model Predictive Control of Input-Affine Systems. The Two Point Boundary Value Problem resulting from the associated Optimal Control Problem is reformulated as an optimization problem, which is locally convex under assumptions coherent with the application. This optimization problem is solved on-line using the gradient descent method, where the gradients are approximated based on geometrical information of the dynamic system differential equations. The resulting control method is summarized in three algorithms. The proposed controller is easy to implement and requires no iterations. As a consequence, the suboptimal control input can be computed in a short time interval, making it ideal for fast highly nonlinear systems. As an example the attitude control of a quadrotor is presented. Simulation results show excellent performance in a wide range of state values, well beyond linear regimes.

This article presents a non-linear control system for a robotic arm that will be used for pipeline reparation. It starts with the mechanical description of the arm, its parameters and how the arm will repair the pipelines. This arm needs a control law that is non-linear. The proposed control strategy is obtained through the interpolation of classic feedback quadratic controllers that are computed for each operation point in a trajectory. This design is based on a numerical model that avoids computing an analytical model which is very useful for this kind of application. The technique is validated through simulations.

This paper presents a high-throughput method for Above Ground Estimation of Biomass (AGBE) in rice using multispectral near-infrared (NIR) imagery captured at different scales of the crop. By developing an integrated aerial crop monitoring solution using an Unmanned Aerial Vehicle (UAV), our approach calculates 7 vegetation indices that are combined in the form of multivariable regressions depending on the stage of rice growth: vegetative, reproductive or ripening. We model the relationship of these vegetation indices to estimate the biomass of a certain crop area. The methods are calibrated by using a minimum sampling area of 1 linear meter of the crop. Comprehensive experimental tests have been carried out over two different rice varieties under upland and lowland rice production systems. Results show that the proposed approach is able to estimate the biomass of large areas of the crop with an average correlation of 0.76 compared with the traditional manual destructive method. To our knowledge, this is the first work that uses a small sampling area of 1 linear meter to calibrate and validate NIR image-based estimations of biomass in rice crops.

The recovery of hand motion is one of the most challenging aspects in stroke rehabilitation. This paper presents an initial approach to robot-assisted hand-motion therapies. Our goal was twofold: firstly, we have applied machine learning methods to identify and characterize finger motion patterns from healthy individuals. To this purpose, Electromyographic (EMG) signals have been acquired from flexor and extensor muscles in the forearm using surface electrodes. Time and frequency features were used as inputs to machine learning algorithms for recognition of six hand gestures. In particular, we compared the performance of Artificial Neural Networks (ANN), Support Vector Machines (SVM) and k-Nearest Neighbor (k-NN) algorithms for classification. Secondly, each identified gesture was turned into a joint reference trajectory by applying interpolation methods. This allowed us to reconstruct the hand/finger motion kinematics and to simulate the dynamics of each motion pattern. Experiments were carried out to create an EMG database from 20 control subjects, and a VICON camera tracking system was used to validate the accuracy of the proposed system. The average correlation between the EMG-based generated joint trajectories and the tracked hand-motion was 0.91. Furthermore, statistical analysis applied to 14 different SVM, ANN and k-NN configurations showed that Fine k-NN and Weighted k-NN have a better performance for the classification of gestures (98% of accuracy). In a future, the trajectories controlled by EMG signals could be applied to an exoskeleton or hand-robotic prosthesis for rehabilitation.

Stroke is the fourth most common cause of death and can lead complex and long-term disability. In this regard, robotic-based rehabilitation could be an alternative for motion recovery. In this research we study how myoelectric signals (EMG) could be used to identify the fingers/hand motion through pattern recognition techniques. To this purpose, we implemented an experimental protocol on three subject groups: (I) non-stroke without hand impairments, (II) stroke without hand impairments and (III) stroke with hand impairments. The subjects performed a set of hand therapies to improve the range of motion and dexterity. Several methods for feature extraction, ranking and classification from EMG signals were implemented and the performance in the motion identification was compared. Specifically, three ranking methods: Two-sample T-test with feature variances, Separability Index, and the Davies-Boulding Index were used to determine the relevance of the features. As a result, dimensionality reduction was achieved by selecting only 50 features out of 136 with a comparable performance. Also, we compared three different classifiers: LDA, KNN and SVM. On average, the KNN classifier obtained a performance of 0.87 followed by the SVM with 0.82 and LDA with 0.74. Experimental results showed that we are able to identify the hand movements from subjects with a stroke event (group III) with 0.85 of correct classification rate average, which seems a promising approach in robotic-based rehabilitation assistance.

This paper addresses the Nonlinear Model Predictive Control of input-affine systems. Based on the Extended Modal Series Method, an explicit suboptimal control law is deduced that solves the two point boundary problem given by Pontryagin’s Maximum Principle in the related Nonlinear Optimal Control Problem. The deduction of the control law is presented and proven. Then the proposed controller is tested in simulation in the attitude control of a quadrotor, a highly nonlinear and unstable system. The simulation results indicate good performance and near optimal behavior. This work is a background for future development of control techniques for input-affine systems.

In Colombia up to 40% of yield variability is due to the effects of climate variations. Rapid phenotyping methods are needed to properly assess the crop and improve production rates. In this paper, we propose to focus on developing a noninvasive system for speeding up monitoring tasks in rice crops. Unmanned Aerial Vehicles are used to gather multispectral visual information for high-throughput crop monitoring. Geo-referenced digital surface models of the crop are generated based on image mosaicing techniques to allow for the autonomous computation of several vegetative indices. This paper presents the implemented system (hardware and software) and a field report of experiments carried out at different crop growth stages.

In this paper we present an approach for analyzing the inertial effects of changing the wing shape for steering a bat-like robot. Using BaTboT, a robotic platform with massive morphing-wings, we have estimated the generation of pitching and rolling torques, which are directly related to forward and turning maneuvers. Results let us conclude that faster retraction of the wings during the upstroke, and slower extension during the downstroke increase both pitching and rolling torques in about 50% compared to those wingbeats with equal periods for retraction/extension. Also, we determined that the pitch torque generation is proportional to $0.6m^{1/f}$ , whereas the rolling torque is promotional to $0.1m^{1/f}$ , being $m$ the mass of the robot and $f$ the flapping frequency of the wings.

In this article, we address the question of how the flight efficiency of Micro Aerial Vehicles with variable wing geometry can be inspired by the biomechanics of bats. We use a bat-like drone with highly articulated wings using shape memory alloys (SMA) as artificial muscle-like actuators. The possibility of actively changing the wing shape by controlling the SMA actuators, let us study the effects of different wing modulation patterns on lift generation, drag reduction, and the energy cost of a wingbeat cycle. To this purpose, we present an energy-model for estimating the energy cost required by the wings during a wingbeat cycle, using experimental aerodynamic and inertial force data as inputs to the energy-model. Results allowed us determining that faster contraction of the wings during the upstroke, and slower extension during the downstroke enables to reduce the energy cost of flapping in our prototype.

© 2015 © 2015 Taylor & Francis and The Robotics Society of Japan. This paper is focused on the safety of mini UAV (mUAV) systems. It presents the great efforts that are being done in air legislation, including the present and future normative. Nevertheless, considering that the work is not finished yet, a low-level risk analysis concept is introduced. Based on the international regulations, a specific three-step structure for mUAV hazard analysis is presented: identification, assessment, and reduction in a recursive loop that provides a solid architecture for facing the wide range of possible risks., This work has been supported by the Robotics and Cybernetics Research Group at Technique University of Madrid (Spain), and funded under the projects ‘ROTOS: Multi-Robot system for outdoor infrastructures protection’, sponsored by Spain Ministry of Education and Science (DPI2010-17998), and ’Robot Fleets for Highly EffectiveAgriculture and Forestry Management’, (RHEA) sponsored by the European Commission’s Seventh Framework Programme (NMP-CP-IP 245986-2 RHEA). The authors want to thank all the project partners: Agencia Estatal Consejo Superior de Investigaciones Científicas - CSIC (Centro de Automática y Robótica, Instituto de Ciencias Agrarias, Instituto de Agricultura Sostenible), CogVis GmbH, Forschungszentrum Telekommunikation Wien Ltd., Cyberbotics Ltd, Università di Pisa, Universidad Complutense de Madrid, Tropical, Soluciones Agrícolas de Precisión S.L., Universidad Politécnica de Madrid – UPM (ETS IngenierosAgrónomos,ETS Ingenieros Industriales),AirRobot GmbH & Co. KG, Università degli Studi di Firenze, Centre National du Machinisme Agricole, du Génie Rural, des Eaux et des Forèts -CEMAGREF, CNH Belgium NV, CNH France SA, Bluebotics S.A. y CM Srl.

This paper presents an Internet of Things (IoT) approach to Human Activity Recognition (HAR) using remote monitoring of vital signs in the context of a healthcare system for self-managed chronic heart patients. Our goal is to create a HAR-IoT system using learning algorithms to infer the activity done within 4 categories (lie, sit, walk and run) as well as the time consumed performing these activities and, finally giving feedback during and after the activity. Alike in this work, we provide a comprehensive insight on the cloud-based system implemented and the conclusions after implementing two different learning algorithms and the results of the overall system for larger implementations.

This paper proposes an integrated system architecture composed by a custom-designed lightweight Ground Penetrating Radar (GPR) and a novel method for low-altitude autonomous flight called Backstepping+DAF. Simulations and experimental results are carried out to evaluate how steady flight is achieved by means of the proposed flight controller that consequently enables the GPR detection of buried objects with similar materials and morphology of real explosive landmines.

This paper presents an Internet of Things (IoT) approach for monitoring temperature and relative humidity applied to product maintenance in hospitals or pharmaceutical entities. Our goal is to integrate a low-cost and scalable network of smart sensors capable of mapping large areas in real-time. In this article, we provide a comprehensive insight into the technologies that compose the IoT architecture: (i) the node layer composed of Wireless Sensor Network (WSN), (ii) the local management layer of the WSN and (iii) the cloud-based layer for enabling remote monitoring. To the date, our IoT system has been working during (8) months in The Hospital Universitario San Ignacio, a 4th level university hospital located in Bogota, Colombia. Here, we present a field report of this work-in-progress system.

Around 8 to 10 million Ton of rice are required in the following years to be able to supply the demand of the overall population. Analysis and monitoring of rice crops becomes nowadays very important issue for farmers, for ensuring a rice production level to cope this demand. This paper presents simulation results of an algorithm that allows to plan and create 2D maps using the technique of image mosaicking with multiple geo-referenced aerial images (multispectral images in the scope of the project). The planning algorithm is called Image Capture algorithm. It takes into account the area the UAV has to cover, the camera configuration, and the state of the UAV in order to define where to take the pictures to build the mosaic. The algorithm presented in this paper was developed in ROS (Indigo) and simulated in Gazebo. The results of this first approach to the 2D mapping of a rice crop allows to see that using the proposed algorithm, it is possible to automate the process of acquiring the pictures for creating the mosaic, ensuring that all the area of interest is covered. By using this algorithm, pictures will be acquired only in specific areas. Therefore, keeping the storage capacity on-board, under control.

The EMG signals are being used in electronic systems with biofeedback control for tracking and classifying of hand motion. These systems present a challenge in identifying the movement due to the variation of the EMG signals between subjects, therefore different pattern recognition techniques have been implemented to overcome this challenge. In response to the previous problem, the present study compares the performance of both K — means and SVM methods to identify five individual movements of the hand. Therefore two techniques of classification were implemented, the first one consist of classifying the movements individually. while the second classifies all five movements through technic based on decision trees. Also this paper analyses the influence of the signal normalization over the performance of the classification. In general, SVM classifier performed better against K — means in the two tests with the error percentage below 9%.

This work presents our initial approach to remote sensing and precision agriculture for monitoring crop fields in Colombia. Using an autonomous quadcopter UAV (unmanned aerial vehicle) equipped with a multispectral camera onboard, our goal is to provide farmers with an integrated tool for measuring and assessing live green vegetation by assembling a terrain image mosaic based on capturing multispectral images of the terrain. This paper presents how to integrate an UAV-based solution for capturing geo-tagged multispectral imagery and the methods to compute multispectral image mosaics. A field report shows preliminary results of the experiments carried out in crop farms located nearby Bogota, Colombia.

[ES] Las aleaciones metálicas que exhiben una propiedad conocida como efecto de memoria de forma, pertenecen a la clase de materiales inteligentes cuya aplicación más notable en el campo de la robótica se refleja en el uso de actuadores musculares artificiales, ó músculos inteligentes. Estos materiales tienen una estructura cristalina uniforme que cambia radicalmente en función de su temperatura de transición, causando su deformación. Se les denomina materiales inteligentes por la capacidad de recordar su configuración inicial después de recibir dicho estímulo térmico. Este artículo presenta la implementación de un actuador muscular inteligente aplicado en un micro-robot aéreo bio-inspirado tipo murciélago. Esto mamíferos voladores desarrollaron poderosos músculos que se extienden a lo largo de la estructura ósea de las alas, adquiriendo una asombrosa capacidad de maniobra gracias a la capacidad de cambiar la forma del ala durante el vuelo. Replicar este tipo de alas mórficas en un prototipo robótico requiere el análisis de nuevas tecnologías de actuación, abordando los problemas de modelado y control que garanticen la aplicabilidad de este actuador compuesto por fibras musculares de SMAs., [EN] The smart muscles are metal alloys that exhibit a property known as the Shape Memory Effect (SME). These materials are composed by an uniform crystal structure that allows the metal alloy to strain as a function of its transition temperature. They are so-called smart materials because the property of recovering their initial configuration after receiving a thermal stimulus. This article presents the use of a smart muscle actuator applied to a bio-inspired micro aerial bat-like robot. These flying mammals have developed powerful muscles that extend along the bone structure of their wings, providing an amazing level of maneuverability. To mimic that kind of morphing wings using an artificial counterpart requires the analysis of new actuation technologies, addressing issues such as the modeling and control that ensure the applicability of this smart actuators composed by Shape Memory Alloys (SMA).

In this paper, a system that allows applying precision agriculture techniques is described. The application is based on the deployment of a team of unmanned aerial vehicles that are able to take georeferenced pictures in order to create a full map by applying mosaicking procedures for postprocessing. The main contribution of this work is practical experimentation with an integrated tool. Contributions in different fields are also reported. Among them is a new one-phase automatic task partitioning manager, which is based on negotiation among the aerial vehicles, considering their state and capabilities. Once the individual tasks are assigned, an optimal path planning algorithm is in charge of determining the best path for each vehicle to follow. Also, a robust flight control based on the use of a control law that improves the maneuverability of the quadrotors has been designed. A set of field tests was performed in order to analyze all the capabilities of the system, from task negotiations to final performance. These experiments also allowed testing control robustness under different weather conditions. © 2011 Wiley Periodicals, Inc. © 2011 Wiley Periodicals, Inc.

This paper presents a method for analysing and optimizing the design of a compliant transmission mechanism for a flapping-wing aerial vehicle. Its purpose is of minimizing the peak input torque required from a driving motor. In order to maintain the stability of flight, minimizing the peak input torque is necessary. To this purpose, first, a pseudo-rigid-body model was built and a kinematic analysis of the model was carried out. Next, the aerodynamic torque generated by flapping wings was calculated. Then, the input torque required to keep the flight of the vehicle was solved by using the principle of virtual work. The values of the primary attributes at compliant joints (i.e., the torsional stiffness of virtual spring and the initial neutral angular position) were optimized. By comparing to a full rigid-body mechanism, the compliant transmission mechanism with well-optimized parameters can reduce the peak input torque up to 66.0%.

This paper presents a Simultaneous Localization and Mapping (SLAM) architecture applied to the autonomous navigation of an Automated Guided Vehicle (AGV) within a Flexible Manufacturing System (FMS). One of the key factors to integrate an AGV into a FMS is related to pick and place tasks, where the AGV is capable to pick and deliver objects within the FSM. To achieve such task, the AGV must first recognize the environment and then navigate throughout the FSM autonomously. Here, we present how a mobile robot can be used for indoor positioning and mapping of an industrial facility. To this purpose, we have developed an open-source software package based on the Robot Operating System (ROS) platform that can be easily integrated into an AGV such as the KUKA youBot. This article describes the ROS-based metapackage called kuka youbot arch and the experiments carried out for assessing the SLAM arquitecture performance in terms of mapping accuracy and proper navigation within de FSM.

This paper presents a novel system based on the Internet of Things (IoT) to Human Activity Recognition (HAR) by monitoring vital signs remotely. We use machine learning algorithms to determine the activity done within four pre-established categories (lie, sit, walk and jog). Meanwhile, it is able to give feedback during and after the activity is performed, using a remote monitoring component with remote visualization and programmable alarms. This system was successfully implemented with a 95.83% success ratio.

This paper presents the use of a low-cost quadrotor applied for visual detection of landmine-like objects. Nowadays, landmines in Colombia are harmful and even letal artifacts mostly abandoned in rural areas. A percentage of the overall number of landmines are hand-crafted and partially exposed on the terrain's surface so that they can be triggered. Based on that fact, we propose an artificial vision approach as a complementary tool for landmine detection. To this purpose, we have developed an open-source software package based on the Robot Operating System (ROS) platform that can be easily: (i) integrated with a low-cost quadrotor such as the AR.drone parrot 2.0 and (ii) applied for real-time visual detection of landmine-like objects (tuna cans). This article describes the ROS-based package called drone detection and the experiments carried out for assessing the detection performance for different type of terrains and landmine visibility.

Explosive landmines have cost the lives of hundreds in several countries. This paper presents a field report of a low cost unmanned aerial vehicle -ARdrone 2.0- used as a complemented tool for landmine visual detection in rural scenarios. The main contribution of this work is the practical experimentation with an integrated tool. This tool is composed by the ARdrone quadrotor and a base station for monitoring the mission. Based on visual feedback from the onboard camera, algorithms for landmine detection are applied. Experiments have been carried out to quantify the performance of the platform by means of measuring the percentage of the detection at different set of flight speeds and altitudes. The goal is not only to detect fully visible landmines but also those partially buried. It has been observed an effective percentage of the detection over 80% when flying at low altitudes about 1m from the ground at speeds up to 2.2ms − 1. The visual methods introduced herein might enable the ARdrone quadrotor to be used as a low-cost autonomous platform for safe area coverage applied to landmine detection in real scenarios.

New actuation technology in functional or "smart" materials has opened new horizons in robotics actuation systems. Materials such as piezo-electric fiber composites, electro-active polymers and shape memory alloys (SMA) are being investigated as promising alternatives to standard servomotor technology [52]. This paper focuses on the use of SMAs for building muscle-like actuators. SMAs are extremely cheap, easily available commercially and have the advantage of working at low voltages. The use of SMA provides a very interesting alternative to the mechanisms used by conventional actuators. SMAs allow to drastically reduce the size, weight and complexity of robotic systems. In fact, their large force-weight ratio, large life cycles, negligible volume, sensing capability and noise-free operation make possible the use of this technology for building a new class of actuation devices. Nonetheless, high power consumption and low bandwidth limit this technology for certain kind of applications. This presents a challenge that must be addressed from both materials and control perspectives in order to overcome these drawbacks. Here, the latter is tackled. It has been demonstrated that suitable control strategies and proper mechanical arrangements can dramatically improve on SMA performance, mostly in terms of actuation speed and limit cycles.

In this paper, we describe our research on bio-inspired locomotion systems using deformable structures and smart materials, concretely shape memory alloys (SMAs). These types of materials allow us to explore the possibility of building motor-less and gear-less robots. A swimming underwater fish-like robot has been developed whose movements are generated using SMAs. These actuators are suitable for bending the continuous backbone of the fish, which in turn causes a change in the curvature of the body. This type of structural arrangement is inspired by fish red muscles, which are mainly recruited during steady swimming for the bending of a flexible but nearly incompressible structure such as the fishbone. This paper reviews the design process of these bio-inspired structures, from the motivations and physiological inspiration to the mechatronics design, control and simulations, leading to actual experimental trials and results. The focus of this work is to present the mechanisms by which standard swimming patterns can be reproduced with the proposed design. Moreover, the performance of the SMA-based actuators’ control in terms of actuation speed and position accuracy is also addressed.

In this paper, we describe our current work on bio-inspired locomotion systems using a deformable structure and smart materials, concretely Shape Memory Alloys, exploring the possibility of building motor-less and gear-less robots. A swimming underwater robot has been developed whose movements are generated using such actuators, used for bending the backbone of the fish, which in turn causes a change on the curvature of the body. This paper focuses on how standard swimming patterns can be reproduced with the proposed design, using an actuation dynamics model identified in prior work.

Aerial multi-robot systems are a robust remote sensing choice to collect environmental data from the Earth's surface. To accomplish this mission in a collaborative way, unmanned aerial vehicles must perform a full coverage trajectory over a target area while acquiring imagery of it. In this paper we address the multi coverage path planning problem with an aerial vehicles team. The approach proposed is hybrid, since is it is composed by an on-line and an off-line steps. This work is based on an optimal solution which is discretized to compute the coverage paths. This work proposes a multi coverage path planning solution making use of computer graphics tools in the world transformation from continuous to discrete, focusing on the aerial images acquisition. The workspace transformation from continuous to discrete is discussed and raster graphics based algorithms are employed.