This study introduces an innovative smart wheelchair designed to improve mobility and health monitoring for individuals with disabilities. Overcoming the limitations of traditional wheelchairs, this smart wheelchair integrates a tri-wheel mechanism, enabling smooth navigation across various terrains, including stairs, thus providing greater autonomy and flexibility. The wheelchair is equipped with two smart Internet of Things (IoT)-based subsystems for control and vital sign monitoring. Besides a joystick, the wheelchair features an electroencephalography (EEG)-based brain–computer interface (BCI) for hands-free control. Utilizing support vector machine (SVM) algorithms has proven effective in classifying EEG signals. This feature is especially beneficial for users with severe physical disabilities, allowing them to navigate more independently. In addition, the smart wheelchair has comprehensive health monitoring capabilities, continuously tracking vital signs such as heart rate, blood oxygen levels (SpO₂), and electrocardiogram (ECG) data. The system implements an SVM algorithm to recognize premature ventricular contractions (PVC) from ECG data. These metrics are transmitted to healthcare providers through a secure IoT platform, allowing for real-time monitoring and timely interventions. In the event of an emergency, the system is programmed to automatically send alerts, including the patient’s location, to caregivers and authorized relatives. This innovation is a step forward in developing assistive technologies that support independent living and proactive health management in smart cities.

This study reports on the development and implementation of the fuzzy incremental controller on a small‐scale linear fresnel reflector solar plant. The control problem is concerned with forcing the output temperature to follow the reference, despite the existence of disturbances. Owing to the non‐linear effects of solar plants, the capability of fixed parameters of the proportional–integral controller to cope with the control problem is limited and leads to unsatisfactory control performance. The proposed controller is a modification to a conventional proportional–integral algorithm with better tuning flexibility. Firstly, the ant colony optimisation algorithm is used to define the optimal parameter of proportional–integral plus series feed‐forward controller while guaranteeing a satisfactory performance of the plant. Secondly, the resulting controller is replaced with an equivalent fuzzy knowledge‐based controller with the error and change of error of the plant as the input variables and flow rate as an output variable. The proposed controller is tested on a quasi‐dynamic model of linear fresnel reflector plant using conventional and renewable energy optimisation toolbox in the environment of MATLAB/Simulink. The controller's performance is compared to the proportional–integral controller, where its effectiveness is evaluated under nominal conditions, measurement delay, noise, and presence of disturbance.

The main target of optimal motion planning is the generation of a trajectory that satisfies objectives, such as minimizing path traveling distance or time interval or obstacle avoidance, and satisfying the kinematics and dynamics of the robot. This paper presents a feasible method for determining the optimal time trajectory planning of robot manipulators using different degrees of B-spline which provide a continuity of position, velocity, acceleration and jerk. The Sequential Quadratic Programming (SQP) algorithm is used for solving the optimization problem taking into account the limits on the velocities, and accelerations for each joint of the robot. The proposed method is tested for movements of the first three-DOF of the CRS CataLyst-5T robot arm.

State-of-the-art Center of Gravity (CoG) estimation methods often face accuracy limitations due to significant errors introduced by commercial force sensors. Additionally, these methods typically require more than three poses and rely on complex matrices and equations, leading to prolonged computational times. This study introduces an advanced sensor system for precise CoG determination that requires only two poses, integrating a novel 4-axis force sensor with a machine learning (ML) model. The sensor design is optimized based on Timoshenko beam theory and validated through finite element analysis and experimentation, achieving average measurement errors below 0.82% and interference errors under 0.71%, demonstrating its accuracy and reliability. The ML dataset comprises approximately 391,000 data points, utilizing 12 input variables to predict CoG represented by 4 variables. Various tree-based ML models—including Decision Tree, Random Forest, Extra Trees, Extreme Gradient Boosting, and Light Gradient Boosting Machine—were evaluated, with hyperparameter tuning performed using Optuna and Bayesian optimization. The results demonstrated that all the developed ML models effectively captured the underlying patterns in the data, as evidenced by evaluation metrics. Among these, the Extra Trees model outperformed the others in terms of accuracy and reliability, as indicated by its lowest mean absolute error of 0.000 N for weight estimation and 0.009 mm for position estimation. These findings underscore the potential of the proposed system and model combination as a highly efficient and scalable solution for precise CoG estimation, with significant implications for industrial and robotic applications.