This full research paper expands on the results and discussions from a case study conducted within a Brazilian EdTech company. It focuses on the accuracy of AI-based speech recognition models for automatic transcription of video lectures, aiming to improve the accessibility of Learning Objects (LOs), especially the audible ones. Previous work assessed the quality of automatic transcriptions in English, Portuguese, and Spanish from Speech- To- Text (STT) services provided by major market players (Amazon, Google, IBM, Microsoft, and OpenAI), using a quantitative approach based on lexical similarity algorithms. Statistically significant differences were identified between the providers and languages evaluated, indicating that STT service can affect the quality of audible OAs enriched with transcriptions or subtitles. This study introduces a new dimension of analysis by incorporating the perceptions of technology students on the accuracy of automatic transcriptions, gathered through an anonymous survey with 56 participants. The survey captured mainly quantitative data using a Likert scale on the accuracy of transcriptions. To provide a comprehensive analysis, this study employs a data triangulation approach, integrating (1) quantitative data from lexical similarity methods, (2) quantitative data from the survey, and (3) findings from a complementary review of the literature. The qualitative analysis, applying Grounded Theory principles to both the survey data and the bibliographic review, enables the exploration of emerging themes and enriches the understanding of factors influencing the quality of automatic transcriptions. This effort underscores the importance of a user-centered perspective and demonstrates the complexity of evaluating STT technologies, pointing to the necessity for future research that combines quantitative and qualitative methods. Additionally, it highlights the relevance of STT in various educational contexts and the need to align such technologies with principles of accessibility and inclusion. By aiming to create more accessible LOs, this work emphasizes the need to develop solutions that ensure inclusion and equity in educational access, reflecting a concerted effort to meet the diverse needs of learners and promote a more welcoming and inclusive learning environment.

In this paper, we extend the Monte Carlo Localization formulation for a more efficient global localization using coarse digital maps (for instance, the OpenStreetMap maps). The proposed formulation uses the map constraints in order to reduce the state dimension, which is ideal for a Monte Carlo-based particle filter. Also, we propose including to the data association process the matching of the traffic signals’ information to the road properties, so that their exact position do not need to be previously mapped for updating the filter. In the proposed approach, no low-level point cloud mapping was required and neither the use of LIDAR data. The experiments were conducted using a dataset collected by the CARINA II intelligent vehicle and the results suggest that the method is adequate for a localization pipeline. The dataset is available online and the code is available on GitHub.

In the path planning task for autonomous mobile robots, robots should be able to plan their trajectory to leave the start position and reach the goal, safely. There are several path planning approaches for mobile robots in the literature. Ant Colony Optimization algorithms have been investigated for this problem, giving promising results. In this paper, we propose the Max-Min Ant System for Dynamic Path Planning algorithm for the exploratory path planning task for autonomous mobile robots based on topological maps. A topological map is an environment representation whose focus is the main reference points of the environment and their connections. Based on this representation, the path can be composed by a sequence of state/actions pairs, which facilitates the navigability of the path, with no need to have the information of the complete map. The proposed algorithm was evaluated in static and dynamic environments, showing promising results in both of them. Experiments in dynamic environments show the adaptability of our proposal.

Unmanned Aerial Systems (UASs) are being employed in many different applications, and with the increased usage security concerns become more accentuated. Communication links are core components of a UAS, and providing a secure communication channel is a necessity in which Authentication strategies play a major role. However, resource-constrained devices cannot adopt off-the-shelf security strategies that were not developed taking into account those devices' constraints. Because of that, new security strategies, including authentication, were purposed for these applications. In this paper, we analyze and compare two authentication protocols developed for WSNs and modified for UAV usage. Tests were performed analyzing time execution and CPU usage on security-specific operation such as hash tables and elliptic curve operations.

The present work demonstrates how an autonomous navigation system of `End-to-End' deep learning principles is directly improved in its response process, depending on the information obtained by different input images configurations. For this, a methodology was developed to allow working with RGB and depth images, which were obtained through a Microsoft Kinect V2 sensor device. Three cameras were used for this experiment. The images of the different cameras were concatenated or grouped, generating new and different input configurations from the vision system. To develop the presented methodology, two support and validation systems were implemented. Through the process of computer simulation, it was able to test the first approaches and define the most important ones. In order to validate the proposed methodology and solutions in real world situations, a 1/4 scale automotive vehicle was prototyped. Finally, the experiments shows the importance of the use of multi-cameras systems for a better performance of autonomous navigation systems based on End-to-End learning approach, heaving an average error of 2.41 degrees in the best configuration tested, with three RGB cameras.

Interaction among vehicles based on wireless communication technologies is an efficient way to improve traffic conditions and safety. In this context, The Grand Cooperative Driving Challenge (GCDC) is an event that aims at developing cooperative systems for intelligent vehicles. Although the two editions of the event (2011 and 2016) have provided conditions for a series of experiments such as platoon stability and lane-merging, the traffic flow in lane-merging scenarios didn't receive so much attention. This paper presents a preliminary study on traffic flow using GCDC 2016 interaction protocol for lane-merging by varying the driver imperfection model and the number of vehicles. We also propose an extension of the standard protocol. We evaluated three lane-merging interaction protocols using the Simulation of Urban Mobility (SUMO). The first protocol is very similar to the GCDC 2016, the second is the default lane change model commonly used in SUMO and the third is a variation of the GCDC interaction protocol. The experiments were conducted using similar conditions and number of vehicles of the event. We analyzed the lane-merging total time for all vehicles to complete the maneuver, the maximum string platoon length and the average platoon speed. Therefore, we could observe that the interaction protocol extension proposed decreased the duration time to complete the lane-merging maneuver of all vehicles without drastically compromising maximum string length and average platoon speed. The flexibility of this extension is closer to real traffic scenario, in which vehicles can perform the merge without having a pre-defined order.

The recent advances in computational processing and the low prices of sensors able to capture three-dimensional information have contributed for the progress of computer vision researches involving 3D data and 3D images. Object recognition allows us to develop complex applications for intelligent mobile robotics, augmented reality, systems for the visually impaired, among other applications. In this context, this paper presents a method for recognizing and classifying objects which are represented in three dimensions through depth maps. The data used in this study comes from the "UW RGB-D Object Dataset" from University of Washington, which is available online and is largely used to evaluate 3D object classifiers. This object database is composed of depth maps captured by the Microsoft's Kinect sensor. The obtained results are promising and contribute positively to the computer vision area.

In this work, a 3D vision system has been developed using a frontal Kinect v2 sensor to monitor the driver, enabling to recognize the use of a cell phone while driving, avoiding driving risks. In fact, when cars are driven by people on phone calls, it increases between 4 and 6 times the risk of crash. The Naturalistic Driver Behavior Dataset (NDBD) was created specifically for this work and it was used to test the proposed system. The proposed solution uses two analysis of the driver's hands positions, the Short-Term (ST) and Long-Term (LT) pattern recognition subsystems, thus it could detect the cell phone usage by the driver in hand-held situations. The system has 3 levels of alarm: no alarm, lowest alarm, and highest alarm. ST detects between no alarm or some level alarm. LT is responsible for determining the risk alarm level, low or high. The classifiers are based on Machine Learning and Artificial Neural Nets (ANN), furthermore, the values set to adjust input features, neuron activation functions, and network topology/training parameters were optimized and selected using a Genetic Algorithm. The best system performance results obtained in the experiments achieved 95% of accuracy in NDBD frames. For the ST classifier, it was used length periods of 5 frames and a window of 80 or 210 frames for LT. The best results achieved obtained only 1% of “no risk” situation having a wrong prediction (false positives with alarm activation), contributing to the driver comfort when he/she is using the system.

We developed a bioinspired algorithm to assist in the navigation of an autonomous mobile robot in dynamic environments. The robotic controller uses both an Artificial Neural Network (ANN) and a Genetic Algorithm (GA), aided by a low computational cost vision system. The controller uses the vision system and the ANN to detect and recognize obstacles found in the robot's path. If the object is in the controller's knowledge bank a previously registered deviation solution is applied. Otherwise, the GA must optimize a new route alternative. We modeled and simulated the controller using robot's simulator V-REP, and the Computer Vision System using Scilab software. The contribution of this paper is the development of a hybrid neuro-genetic algorithm to control the navigation of autonomous mobile robots in dynamic environments.

This paper presents a system for detection and recognition of traffic signs using 2D images and 3D scene data. The detection and recognition of the 3D structures (pole and signs) and the classification of the traffic signs is based on a new 3D features extraction method and the use of a Deep Learning method. The traffic sign recognition aims to increase road safety considering autonomous and semi-autonomous intelligent robotic vehicles in structured environments with traffic rules (urban streets or roads/highways). The proposed system can be used as an Advanced Driver Assistance System (ADAS) to help human drivers to increase safety and to respect the traffic rules while driving the car. It can also be adopted in fully autonomous vehicles, in the task of detecting traffic signs, making it possible to adapt the vehicle navigation control according to the local traffic rules. The system must be able to detect traffic signs, using 3D data in point clouds, and classify several different traffic signs, using 2D data considering colors, textures and shapes information (e.g. maximum speed allowed, stop, slow down, turn ahead, pedestrian crossing). The results are promising and very satisfactory, we obtained an accuracy of 97.64% in the 2D classification task and 76% accuracy in the single frame 3D detection task. These results were obtained using for testing a well-known dataset of street scenes with traffic signs, The KITTI Vision Benchmark Suite, and also another traffic sign benchmark dataset, the INI - German Traffic Sign Benchmark.

In the path planning problem for autonomous mobile robots, robots have to plan their path from the start position to the goal. In this paper, we investigate the application of the MMAS algorithm to the exploratory path planning problem, in which the robots should explore the environment at the same time they plan the path. Max-min ant system is an ant colony optimization algorithm that exploits the best solutions found. In addition, to analyze the quality of solutions obtained, we also analyze the traveled distance spent by robots in the first iteration of the algorithm. The environment is previously unknown to the robots, although it is represented by a topological map, that does not require precise information from the environment and provides a simple way to execute the navigation of the path. Thus, the paths are represented by a sequence of actions that the robots should execute to reach the goal. The navigation of the best solution found was implemented in a realistic robotic simulator. The proposed algorithm provides a very good performance in relation to a genetic algorithm and the well-known A∗ algorithm that deal with this problem.

In this work, a vision system has been developed using a frontal camera to monitor the driver, enabling to recognize the use of a cell phone while driving. It is estimated that 80% of car crashes and 65% of near collisions involved drivers who were inattentive in traffic for three seconds before the event. Five videos in real environments were generated to test the proposed system. The solution is a hybrid system and that uses a pattern recognition system (PR) for classification and a movement detection system (MD) for choosing the PR parameters at the end of each period of 3 seconds. The PR parameters are the threshold (frames identified as a cell phone use) and classifier selection. The classifiers are based on ANN, furthermore, the value of constants in neuron activation function and network training parameters were adopted with a genetic algorithm. Experimentally, it was established that when the movement indicates a possible use of the cell phone, the threshold 60% and an MLP/Gaussian classifier with seven neurons in intermediate layer are suitable; otherwise, a threshold of 85%, and MLP/Gaussian with two neurons in intermediate layer for classification are used. The average accuracy achieved was 91.68% in real environment scenes.

Detecting changing traffic conditions is of primal importance for the safety of autonomous cars navigating in urban environments. Among the traffic situations that require more attention and careful planning, road junctions are the most significant. This work presents an empirical study of the application of well known machine learning techniques to create a robust method for road junction detection. Features are extracted from 3D pointclouds corresponding to single frames of data collected by a laser rangefinder. Three well known classifiers-support vector machines, adaptive boosting and artificial neural networks-are used to classify them into “junctions” or “roads”. The best performing classifier is used in the next stage, where structured classifiers-hidden Markov models and conditional random fields-are used to incorporate contextual information, in an attempt to improve the performance of the method. We tested and compared these approaches on datasets from two different 3D laser scanners, and in two different countries, Germany and Brazil.

One very important component of Hardware implementations of Artificial Neural Networks (HNN) is the activation function. A common solution for activation function design in HNN is the Look-Up Table (LUT) representation that has acceptable area constraints and usually achieves acceptable execution time. Tools and techniques for automatic generation of LUTs have been developed but have some problems as high complexity, high dependency of external tools or languages and the optimization techniques that are not efficient enough for critical applications. In order to solve some of the presented problems this paper describes a method for automatic LUT generation that overcomes some deficiencies of existing methods using a simple approach. This method was evaluated initially considering the xor function and the sine function for evaluation purposes. After these initial tests, two other ANN based robotic systems were tested. These robotic systems will be implemented in hardware in the near future and they demonstrate a different behavior using different domains of the same activation function. Proposed method achieved good results that are directly related to how neural networks define the activation function domain.

The path planning task for mobile robots consists of define a trajectory to the robot leaves its start position and reach its goal without to collide with obstacles. In general, the robot needs to know previous information about the environment (e.g. maps, predefined routes) to plan its trajectory. In an exploration task, the robot does not know the environment and discovers it when moving to reach the goal coordinates. In this paper, an exploratory path planning aiming to reach a goal position is studied and a new method based on genetic algorithm, topological environment representation and realistic robot actions is proposed. In this method, the robots execute a sequence of reliable local actions (simple reactive behaviors) to move through the unknown environment, adopting a topological environment representation. They plan the path at the same time the environment is explored, in which the genetic algorithm evolves the sequence of actions to be executed. The results show that the squad of robots (GA population) reach the goal faster than an individual search. The proposed approach deal with environment traps better than the classical search A* algorithm and a variation of the A*, named C*, here also introduced.

Robust and stable control is a requirement for navigation of self-driving cars. Some approaches in the literature depend on a high number of parameters that are often difficult to estimate. A poor selection of these parameters often reduces considerably the efficiency of the control algorithms. In this paper we propose a simplified control system for autonomous vehicles that depends on a reduced number of parameters that can be easily set. This control system is composed of longitudinal and lateral controllers. The longitudinal controller is responsible for regulating the vehicle's cruise velocity while the lateral controller steers the vehicle's wheels for path tracking. Simulated and experimental tests have been carried out with the CaRINA II platform in the university campus with positive results.

The detection of a traffic light on the road is important for the safety of persons who occupy a vehicle, in a normal vehicles or an autonomous land vehicle. In normal vehicle, a system that helps a driver to perceive the details of traffic signals, necessary to drive, could be critical in a delicate driving manoeuvre (i.e crossing an intersection of roads). Furthermore, traffic lights detection by an autonomous vehicle is a special case of perception, because it is important for the control that the autonomous vehicle must take. Multiples authors have used image processing as a base for achieving traffic light detection. However, the image processing presents a problem regarding conditions for capturing scenes, and therefore, the traffic light detection is affected. For this reason, this paper proposes a method that links the image processing with an estimation state routine formed by Hidden Markov Models (HMM). This method helps to determine the current state of the traffic light detected, based on the obtained states by image processing, aiming to obtain the best performance in the determination of the traffic light states. With the proposed method in this paper, we obtained 90.55% of accuracy in the detection of the traffic light state, versus a 78.54% obtained using solely image processing. The recognition of traffic lights using image processing still has a large dependence on the capture conditions of each frame from the video camera. In this context, the addition of a pre-processing stage before image processing could contribute to improve this aspect, and could provide a better results in determining the traffic light state.

An autonomous car must have a robust perception system to navigate safely in urban streets. An important issue of environment perception is the road (navigable area) detection and the identification of the road geometry. The road geometry information can be used to determine the vehicle control according to the street and also for topological localization. Existing road geometry identifiers only work with a limited number of classes and, due to the use of cameras, some solutions depend on filters to deal with shadows and light variations. This paper presents a road detector that extracts curb and navigable surface information from a multilayer laser sensor data. The road data was trained with an artificial neural network (ANN) and classified into eight road geometries: straight road, left turn, right turn, left side road, right side road, T intersection, Y intersection and crossroad. The main advantage of our method is its robustness to light variations for detecting distinct roads even in the presence of noisy data thanks to the ANN. In order to determine which road information has the best features for ANN training, three approaches were explored: ANN trained with curb data, ANN trained with surface data and ANN trained with both curb and surface data. Performed experiments resulted in the superiority of the network trained with both curb and surface data, with an accuracy of 0.91799. The trained ANN was validated in different urban scenarios and, evaluating a 1 Km track, we obtained a 94.48% of correct classifications. These results are superior than other works that detect fewer number of road shapes.

Curb detection is an important capability for autonomous ground vehicles in urban environments. It is particularly useful for path planning and safe navigation. Another important task that can benefit from curb detection is localization, which is also a major requirement for self-driving cars. There are several approaches for identifying curbs using stereo cameras and 2D LIDARs in the literature. Stereo cameras depend on image pair matching methods to obtain depth information. Although 2D LIDARs being able to directly return this information, only few curb points can be detected using this sensor. In this work we propose the use of a 3D LIDAR which provides a dense point cloud and thus make possible to detect a larger extent of the curb. Our approach introduces the use of robust regression method named least trimmed squares (LTS) to deal with occluding scenes in contrast of temporal filters and spline fitting methods. We also used the curb detector as an input of a Monte Carlo localization algorithm, which is capable to estimate the pose of the vehicle without an accurate GPS sensor. We conducted experiments in urban environments to validate both the curb detector and the localization algorithm. Both method delivered successful results in different traffic situations and an average lateral localization error of 0.52655 m in a 800 m track.

An accurate and efficient method for obstacle detection is a key component of a robotic navigation system. Concerning indoor environments, the ground surface can be modeled as a plane (or a set of) and once estimated it can be employed for obstacle detection, e.g. points lying above and below are considered obstacles. The same does not hold for off-road and urban scenarios where one cannot expect planar surfaces or obvious structural patterns. In 2002, Talukder et al. presented a method to deal with such environments. Their method is based on the height difference and “slope” between three-dimensional points. Despite having been used successfully on several occasions, the method has a high computational cost. We propose the use of a Graphics Processing Unit (GPU) to enable its execution in real time. Experiments were performed using a stereo camera and an RGB-D sensor, where the GPU implementation has been compared to multi-core and single-core CPU implementations. The results reveal a significant gain in computational performance, reaching a speedup of almost 80 times in a specific instance.

The application of pesticides and fertilizers in agricultural areas is of prime importance for crop yields. The use of aircrafts is becoming increasingly common in carrying out this task mainly because of its speed and effectiveness in the spraying operation. However, some factors may reduce the yield, or even cause damage (e.g. crop areas not covered in the spraying process, overlapping spraying of crop areas, applying pesticides on the outer edge of the crop). Climatic conditions, such as the intensity and direction of the wind while spraying add further complexity to the control problem. In this paper, we describe an architecture based on unmanned aerial vehicles (UAVs) which can be employed to implement a control loop for agricultural applications where UAVs are responsible for spraying chemicals on crops. The process of applying the chemicals is controlled by means of the feedback obtained from the wireless sensor network (WSN) deployed on the crop field. The aim of this solution is to support short delays in the control loop so that the spraying UAV can process the information from the sensors. We evaluate an algorithm to adjust the UAV route under changes in wind intensity and direction. Moreover, we evaluate the impact of the number of communication messages between the UAV and the WSN. Results show that the adjustment of the route based on the feedback information from the sensors could minimize the waste of pesticides.

The development of autonomous vehicles is a highly relevant research topic in mobile robotics. Road recognition using visual information is an important capability for autonomous navigation in urban environments. Over the last three decades, a large number of visual road recognition approaches have been appeared in the literature. This paper proposes a novel visual road detection system based on multiple artificial neural networks that can identify the road based on color and texture. Several features are used as inputs of the artificial neural network such as: average, entropy, energy and variance from different color channels (RGB, HSV, YUV). As a result, our system is able to estimate the classification and the confidence factor of each part of the environment detected by the camera. Experimental tests have been performed in several situations in order to validate the proposed approach.

This paper presents the technical details of Tiriba, which is a small autonomous electrical airplane used into pre-defined missions and applications such as agricultural and environmental monitoring. This class of aircrafts has taken advantage of the miniaturization of electronic components and modules such as GPS receivers, digital cameras, wireless communication equipment, and other sensors. One of the main components of Tiriba is the autopilot module, which is capable to control the aircraft and keep it in a predefined route. This paper addresses the use of model driven development of the system and its multiprocessor platform to support the four components that compose the UAV control software architecture.

This paper describes our studies in the legged robots research area and the development of the LegGen System, that is used to automatically create and control stable gaits for legged robots into a physically based simulation environment. The parameters used to control the robot are optimized using genetic algorithms (GA). Comparisons between different robot models and fitness functions were accomplished, indicating how to compose a better multi-criterion fitness function to be used in the gait control of legged robots. The best gait control solution and the best robot model were selected in order to help us to build a real robot. The results also showed that it is possible to generate stable gaits using GA in an efficient manner.

The main goal of this paper is to describe our research and implementation of a handwritten signature authentication system based on artificial neural networks. In this system the authentication process occurs in the following way: firstly, the users' signatures are read using a pen tablet device and then stored; after that some adjustments in position and scale are accomplished; representative signature features are extracted; the input space dimensionality is reduced using principal component analysis; and finally, the users' signatures are classified as authentic or not, through the use of a neural network. Several experiments were accomplished using a 2440 real signatures database, and the obtained results were very satisfactory.

This paper describes the simulation system proposed in order to study and to implement intelligent autonomous vehicle control. The developed system can automatically drive a vehicle, implementing a robust control system capable of simulating in a realistic way autonomous parking in a parallel parking space. The system controls the vehicles based on the reading of sonar sensors and uses a neural network to automatically generate acceleration and steering commands, parking it in a parallel parking space. The controller was implemented using a Jordan-Net based neural network, and the results obtained in our simulations demonstrated that the proposed controller is perfectly able to correctly park the vehicle in different situations.

This paper presents an intelligent and adaptive virtual environment, which has its structure and presentation customized according to users' interests and preferences (represented in a user model) and in accordance with insertion and removal of contents in this environment. An automatic content categorization process is applied to create content models, used in the spatial organization of the contents in the environment. An intelligent agent assists users during navigation in the environment and retrieval of relevant information. In order to validate our proposal, a prototype of a distance learning environment, used to make educational content available, was developed.

This work presents an intelligent virtual environment, called AdapTIVE (adaptive three-dimensional intelligent and virtual environment), which has its structure and presentation customized according to users' interests and preferences (represented in a user model) and in accordance with insertion and removal of contents in this environment. An automatic content categorization process is applied to create content models, used in the spatial organization of the contents in the environment. An intelligent agent assists users during navigation in the environment and retrieval of relevant information. This is a promising approach for new and advanced forms of education, entertainment and e-commerce. In order to validate our approach, a case study of an e-commerce environment is presented.