18 September 2023 in IEEE Transactions on Intelligent Vehicles
An autonomous vehicle must be able to locate itself precisely and reliably in a large-scale outdoor area. In an attempt to enhance the localization of an autonomous vehicle based on Global Navigation Satellite System (GNSS)/Camera/Inertial Measurement Unit (IMU), when GNSS signals are interfered with or obstructed by reflected signals, a multi-step correction filter is used to smooth the inaccurate GNSS data obtained. The proposed solutions integrate a high amount of data from several sensors to compensate for the sensors' individual weaknesses. Additionally, this work proposes a Generalizable Deep Visual Intertial Odometry (GD-VIO) to better locate the vehicle in the event of GNSS outages. The algorithms suggested in this research have been tested through real-world experimentations, demonstrating that they are able to deliver accurate and trustworthy vehicle pose estimation.
18 February 2026 in Scientific Reports
This study proposes a robust cascaded hybrid control architecture combining the Linear Quadratic Regulator (LQR) with Super-Twisting Sliding Mode Control (STSMC) to address the attitude stabilization problem of underactuated quadrotor UAVs. In this framework, the inner LQR loop ensures optimal stabilization, while the outer STSMC loops provide robust reference modification to compensate for uncertainties. To maximize the robustness of the proposed hybrid architecture, the parameters of the STSMC component were globally optimized using the Big Bang-Big Crunch (BB-BC) algorithm. The effectiveness of the proposed strategy is rigorously evaluated on the Quanser 3-DOF Hover system model against two distinct benchmarks: the classical LQR and the optimization-based Model Predictive Control (MPC). Comprehensive simulation scenarios were conducted, including stochastic wind disturbances, time varying thrust coefficients, sudden payload release, and critical motor failure. Additionally, a virtual 3D spatial analysis based on kinematic reconstruction was performed to validate the positional stability of the rotor tips. The comparative results demonstrate that while the MPC controller exhibits a faster transient response compared to the classical LQR, it suffers from significant performance degradation under model mismatches. In contrast, the proposed LQR-STSMC consistently outperforms both benchmarks, achieving error metrics (ISE, IAE, and ITAE) orders of magnitude lower than the competitors. The findings confirm that the hybrid scheme offers superior fault tolerance, faster stabilization, and precise tracking capabilities, making it a highly reliable alternative for safety critical flight operations in uncertain environments.
9 January 2026 in Physica Scripta
This study presents a comparative analysis of different control methodologies on the linearized model of a single degree-of-freedom (1-DOF) helicopter system around its operating point. The control structures evaluated in this study include a Proportional-Integral-Derivative (PID) controller optimized via Particle Swarm Optimization (PSO), Model Predictive Control (MPC), and the Deep Deterministic Policy Gradient (DDPG) algorithm based on Deep Reinforcement Learning (DRL). The conducted simulations revealed that the DRL-based controller exhibited a superior performance compared to the other methods, even on the linearized model, both in terms of reducing the magnitude of the error and improving the system’s transient response performance. This approach demonstrates the capacity to produce a rapid response without compromising the system’s stability, concurrently achieving faster rise and settling times. Furthermore, by effectively suppressing the time-weighted error values, it provides an advantage in terms of both accuracy and control quality. These compelling results indicate the significant potential of learning-based control methods in the control of nonlinear systems, establishing DRL-based strategies as a reliable foundation for advanced control applications and a superior alternative to classical methods.
1 September 2022 in Tehnički glasnik
Rubber metal bushings are one of the most important parts of the suspension system. These parts are specially designed to minimize the vibrations and noises that vehicles are exposed to while driving. In this study, it is aimed to measure the circularity of rubber metal bushings after stress relief. A modular design is provided according to the tolerance values of the bushings in the inventory. Simultaneously with deep learning and image processing algorithms, laser circularity measurements are verified with reference to bush circularity and deformation points are determined. The detected deformation is transferred to the HMI screen. With this study, it was ensured that the circularity errors in the rubber metal bushings were taken under control and the problems that may be encountered after the production were eliminated, a more sensitive solution was provided, and a quality production and accurate results were obtained
31 August 2022 in Turkish Journal of Forecasting
Unmanned underwater vehicles (ROV/AUV) are autonomous or remotely controlled robotic systems that can move underwater at any desired angle. Unmanned underwater vehicles; It is used in areas such as underwater image taking, ship maintenance and repair, coast guard, examination of shipwrecks, underwater cleaning. In this study, the software design of the balance control of underwater vehicles was carried out using the PID algorithm. For the PID algorithm trial, a two-motor test setup with an IMU sensor was prepared. After the data from the sensor were recorded in MATLAB using the Kalman filter, the transfer function of the system was obtained using the System Identification Toolbox. With the obtained transfer function, the stable operation of the system is provided in real time. As a result of the research on software and hardware integration, microcontroller ARM-based STM32 was used.