Visassist

Autonomous vehicle research often requires expensive hardware platforms equipped with multiple sensors, which limits accessibility for students and early-stage researchers. To address this challenge, this paper presents the design and implementation of a low-cost autonomous robot car using a Raspberry Pi integrated with an AI-based voice assistant. The proposed system enables autonomous navigation, obstacle detection, fog-aware vehicle detection, multilingual voice control, and automatic parking. Sensor fusion is achieved using ultrasonic sensors, camera modules, and optional thermal sensing to ensure safe operation in low-visibility conditions. Voice commands primarily in English are processed using a cloud-based speech recognition service, allowing natural human–vehicle interaction. The system architecture is modular, enabling gradual sensor integration and future scalability. Experimental results demonstrate reliable obstacle avoidance, accurate voice command execution, and successful autonomous parking, validating the effectiveness of the proposed platform as a practical testbed for intelligent transportation research.The rapid development of autonomous driving technologies has emphasized the need for affordable and scalable experimental platforms capable of validating perception, decision-making, and control algorithms in real-world conditions.

Children with profound hearing loss can gain access to sound through cochlear implants (CIs), but these devices must be worn consistently to promote auditory development. Although subjective parent reports have identified several factors limiting long-term CI use in children, it is also important to understand the day-to-day issues which may preclude consistent device use. In the present study, objective measures gathered through datalogging software were used to quantify the following in children: (1) number of hours of CI use per day, (2) practical concerns including repeated disconnections between the external transmission coil and the internal device (termed "coil-offs"), and (3) listening environments experienced during daily use. This study aimed to (1) objectively measure daily CI use and factors influencing consistent device use in children using one or two CIs and (2) evaluate the intensity levels and types of listening environments children are exposed to during daily CI use. Retrospective analysis. Measures of daily CI use were obtained from 146 pediatric users of Cochlear Nucleus 6 speech processors. The sample included 5 unilateral, 40 bimodal, and 101 bilateral CI users (77 simultaneously and 24 sequentially implanted). Daily CI use, duration, and frequency of coil-offs per day, and the time spent in multiple intensity ranges and environment types were extracted from the datalog saved during clinic appointments. Multiple regression analyses were completed to predict daily CI use based on child-related demographic variables, and to evaluate the effects of age on coil-offs and environment acoustics. Children used their CIs for 9.86 ± 3.43 hr on average on a daily basis, with use exceeding 9 hr per day in ∼64% of the children. Daily CI use reduced significantly with increasing durations of coil-off (p = 0.027) and increased significantly with longer CI experience (p < 0.001) and pre-CI acoustic experience (p < 0.001), when controlled for the child's age. Total time in sound (sum of CI and pre-CI experience) was positively correlated with CI use (r = 0.72, p < 0.001). Longer durations of coil-off were associated with higher frequency of coil-offs (p < 0.001). The frequency of coil-offs ranged from 0.99 to 594.10 times per day and decreased significantly with age (p < 0.001). Daily CI use and frequency of coil-offs did not vary significantly across known etiologies. Listening environments of all children typically ranged between 50 and 70 dBA. Children of all ages were exposed to speech in noisy environments. Environment classified as "music" was identified more often in younger children. The majority of children use their CIs consistently, even during the first year of implantation. The frequency of coil-offs is a practical challenge in infants and young children, and demonstrates the need for improved coil retention methods for pediatric use. Longer hearing experience and shorter coil-off time facilitates consistent CI use. Children are listening to speech in noisy environments most often, thereby indicating a need for better access to binaural cues, signal processing, and stimulation strategies to aid listening. Study findings could be useful in parent counseling of young and/or new CI users.

El sistema educativo contemporáneo está influenciado por los movimientos migratorios y el desarrollo de las Tecnologías de la Información y la Comunicación (TIC). Este trabajo analiza la influencia de las TIC en el rendimiento escolar de estudiantes vulnerables. Para ello, se estudian las relaciones del rendimiento escolar con el país de origen del estudiante, la conexión a Internet en el hogar y el uso diario de dispositivos digitales. Apoyado en la metodología de investigación en educación comparada, se contrastan las muestras representativas de las poblaciones de España, los países miembros de la Unión Europea y la Organización para la Cooperación y el Desarrollo Económicos. Se utilizan fundamentalmente los datos publicados por el Ministerio de Educación en los informes PISA 2018 y OCDE 2019. La estadística descriptiva sustenta los resultados obtenidos. Se analizan las variables participación diaria en redes sociales, participación diaria en juegos online, lectura diaria de noticias online y la nota media de Ciencias y Matemática respectivamente. Los hallazgos muestran que en España existe un bajo rendimiento escolar relacionado con el origen del estudiante; la conexión a Internet en el hogar y que un uso diario de dispositivos digitales puede ser beneficioso, si las exposiciones son moderadas y controladas.

In this paper, we propose a new disparity-based obstacle detection method for intelligent vehicle. The disparity maps captured by the stereo vision system ZED is utilized to detect obstacles in outdoor environment. This method consists of several steps namely, histogram calculation based on the disparity map, peaks finding using Non-Maximum Suppression (NMS), disparity slicing and obstacle detection using contours. The first step is to obtain the distribution of the pixels in the disparity map. The second step is performed to get the peaks in the histogram after the binary processing on the disparity map. However, there are always many noise objects in the binary image. The real obstacle area needs to be separated from the noise obstacles. Thus, the third step is performed to filter the binary image based on a statistics by column and then segment the detected obstacle area. The last step is to calculate the obstacle including its size, distance and position on the disparity map. A series of experiments prove that the proposed method is accurate at detecting obstacles and calculating the required information based on stereovision techniques.

An Intelligent Wheelchair Using Situation Awareness and Obstacle Detection

Recent advances in innovation have made ordinary people's lives richer, simpler, and easier. Many people suffer from vision impairment, which causes many difficulties in their daily work, according to World Health Organization (WHO) statistics. Our goal is to develop a framework that is unobtrusive, safe, portable, and versatile in order to assist blind people in their daily lives. It intends to develop an effective system for assisting the visually impaired with obstacle detection and scene classification. The Raspberry-Pi 3B+ and a camera module are used in the proposed method. It photographs the scene and then preprocesses the images with Viola-Jones and Yolov5 object detection algorithms. The aforementioned techniques are used to detect people and objects. We also used a sound system that speaks when an object or person is detected. The presented study implements simple computations and detects obstacles with high efficiency. Unlike other frameworks, this framework is lightweight and easy to transport

The Research on Autonomous vehicles is profoundly increasing in popularity. Autonomous vehicles are widely preferred for its advanced technological implementation in Automatic emergency breaking systems and forward collision warning. Another important aspect of self-driving cars is that providing driving assistance hence ensuring a safe and flexible driving experience and preventing catastrophic accidents due to drivers’ negligence. Thus proposed a feasible neural network code for obstacle detection and train the model for distance predictions and automatic braking. The proposed system is designed by using Node MCU ESP8266 which receives the data from the external environment using sensors such as camera module and ultrasonic sensor. The Convolutional Neural Network (CNN) processes these data and provides the required breaking based on the distance from the obstacle detected. The camera module sends signals which are used for object classification which is performed by the trained Convoluted Neural Network model. According to the proposed system, the autonomous vehicle successfully predicts the safe driving distance with the trained CNN model in MATLAB R2022b.

Building situational awareness systems that can collect and evaluate data from several sources in real time is crucial as cyber threats become more complex. Sensor fusion, probabilistic reasoning, graph theory, and information theory are all brought together in the innovative mathematical model known as Multi-Source Fusion-Based Cybersecurity Situational Awareness (MF-CSSA). This model is designed to give a complete framework for the visualization and awareness of real-time cybersecurity scenarios. Scalability, accuracy, reaction time, false positive rate, and situational awareness score are the five key performance indicators (KPIs) that we used in order to simulate and assess the proposed model’s architecture. Scalability is the level of scalability that the architecture has. A hundred iterations of the simulation were carried out, and the results indicate that MF-CSSA regularly outperforms the baseline techniques. By combining data and using probabilistic reasoning, the model is able to predict hostile activities in a consistent manner, as shown by the fact that it has an average accuracy of 88% when detecting threats. Having the ability to function in real time is very important, as shown by the fact that its average reaction time is 2.1 seconds. By doing so, it stops attacks from happening before they ever happen. By achieving an average score of 0.82 on the situational awareness scale, the model demonstrates that it is capable of recognizing and contextualizing cybersecurity scenarios. Another way of putting it is that it is able to swiftly adjust to new dangers. The model has a low false positive rate of just 8%, which is in addition to considerably decreasing the amount of work that analysts have to put in and alert fatigue. Last but not least, MF-CSSA is able to develop and handle a significant amount of traffic since it can process around 110 events per second. Because of this, it is a good option for huge networks that experience significant levels of information flow. It is clear from these data that the MF-CSSA architecture is a way of real-time cyber threat defense that is not only intelligent but also practical and provides accurate results.

This article aims to show the possible savings in electricity costs in smart building installations with the use of new version of Raspberry Pi 3 model B + as the control unit in intelligent building automation systems. It presents a comparison of the consumption of electricity in two units used in the central control systems, i.e. a small Windows-based computer and a Raspberry microcomputer. The power consumption of these units was measured during the rest period and during standard operations in the intelligent installation system. The conducted measurements proved that the use of the new updated version of Raspberry Pi 3 model B + as the central control unit in intelligent building management systems is more economical and energy-saving.

This paper presents a novel unsupervised sensor fusion method to detect artificial objects in outdoor environments. We define artificial objects present in outdoor environments using structure and appearance information: an artificial object is composed of several smooth surfaces with sufficiently extended area and distinctive colors with respect to the environment. Structure information is obtained extracting smooth linked surfaces from 3D range data. Appearance information is computed on image data through a color processing. The problem of fusing these two different kinds of information is addressed through the use of a Bayesian sensor fusion approach. A probability map is built and then clustered with a 2.5D unsupervised self organizing network. This method defines objects according to the distribution of the probability values in the fusion map. The resulting clusters are labeled with their mean probability value representing the confidence the detected regions have of being artificial objects. In order to show the validity of the proposed algorithm some experiments have been performed in real outdoor environments showing promising results.

One challenge faced by sensor fusion strategy designed for sensor networks is to real-time achieve high reliability estimated data with the inaccuracy and fault in raw sensor measurements. In this paper, we present an online outlier clearance technique with low computational complexity and memory usage inspired by the nearest neighbor rule that can identify and remove the spurious data in the distributed multisensor fusion process. The proposed technique is fully localized and thus saves communication overhead as well as has a good extension as deployed nodes increasing. In this technique, we define a weighted average distance-based outlier factor criterion to detect the outlier and replace it with estimated data. With the help of contrast simulations, in which we adopt other two typical techniques, it illustrates that the proposed technique provides better performance in real-time clearing outlier during the sensor fusion process without the prior knowledge of outlier.

Sensor fusion techniques are currently being investigated for potential application to a variety of smart weapon systems. Primary sensor candidates include active and passive millimeter wave as well as active and passive infrared configurations. Improved overall dual-mode sensor performance must be carefully traded off with hardware complexity, packaging challenges, and increased costs associated with the sensor fusion process. An additional factor of significance is the selection of sensor geometry, side-by-side, or common aperture.

Development of sensor data fusion for obstacle detection for electric mini tractor

For assisting blind or visually impaired persons, many computer vision technology has been developed. Some camera based systems were developed to help those people in way finding, navigation and finding daily necessities. The motion of the observer causes all scene object stationary or non-stationary in motion. And hence it is very much important to detect moving object with the moving observer. In this context we have proposed a camera based prototype system for assisting blind person in detection of obstacles by using motion vectors. We have collected dataset of their indoor and outdoor environment and estimated the optical flow to perform object detection. Furthermore we have detected the objects in the region of interest without using costly Depth cameras and sensors. The hardware used in the proposed work is ‘Raspberry Pi 2-B’ and the algorithms used for object detection is performed using MATLAB (for simulation purpose) and Python language.

Robotic systems are becoming increasingly crucial in applications requiring high precision. While a robot can operate using basic sensor feedback under controlled conditions, achieving micro-level accuracy requires more comprehensive data integration, especially in dynamic environments. The fusion of data from a variety of sensors is necessary for improving the positioning accuracy of a robot because the accuracy of one type of sensor is insufficient. The field of micro-positioning presents new challenges and tasks that have been gradually explored in the recent literature published from 2015 to 2025. Micro-positioning is a complex operation that involves factors such as mechanical drift, environmental effects, and sensor signal errors. Hybrid fusion is a sensor fusion technique that combines elements of fusion at different levels. For the effective deployment of robots in such contexts, it is essential to integrate multiple sensors and ensure reliable data fusion between them. This involves the use of different sensors, advanced fusion algorithms, and accurate calibration methods through sensor fusion and sophisticated data processing techniques. This literature review presents an analysis of the sensor data fusion methods for precise robot micro-positioning. The focus is on the investigated sensors, the applied synthesis methods, and the developed algorithms and their practical application to identify the existing gaps for future system improvements. Finally, discussions and conclusions based on the collected ideas are presented.