Oğuz, Yasin; Ünker, Faruk; Cetin, Gokhan

This study proposes a hybrid control approach to improve the stability and adaptability of a two-wheeled gyroscopic combat robot by integrating deep learning with Linear Quadratic Regulator (LQR) control. Due to their inherently unstable nature, two-wheeled robots face challenges maintaining balance, especially when Control Moment Gyroscope (CMG) disk speeds vary, which significantly affect system dynamics. While LQR provides effective stabilization by minimizing a cost function, its performance degrades when disk speeds change. To overcome this limitation, a Neural Network (NN) was trained on 42001 precomputed LQR gain sets optimized for CMG disk speeds ranging from 800 to 5000 rad/s. The trained NN predicts appropriate coefficients in real time, enabling adaptive control under dynamic operating conditions. Simulation results demonstrate that the hybrid NN-LQR controller maintains stability under recoil forces up to 3000 N, where the conventional LQR loses balance about 4–6 seconds. Furthermore, the proposed controller reduces energy consumption by 21.6% during steady-state operation within the first 5 seconds and 10.6% under recoil conditions within the first 3 seconds compared to classical LQR. Although developed for a combat robot, the proposed framework offers a scalable solution for broader robotic applications such as search and rescue, surveillance, and autonomous navigation in complex environments, contributing to the advancement of intelligent control in robotics.

Oğuz, Yasin; Çetin, Gökhan; Ünker, Faruk

Two-wheeled gyroscopic robots are challenging to balance due to their intrinsic instability and sensitivity to external disturbances. This study proposes a hybrid control approach that combines a linear quadratic regulator (LQR) and a regression-based machine learning model to increase stability and adaptability. The regression model is trained with data collected from an LQR-controlled system to predict the optimal control responses under different conditions. The proposed approach utilizes state-space representation and the Jacobian linearization technique to design and simulate the robot’s performance under both controlled and random disturbance profiles. Simulation results indicate that the proposed method ensures steady motion even in dynamic, uncertain environments and increases resistance to external disturbances. Compared to the conventional LQR control method, the regression-enhanced model dynamically adjusts the control parameters, effectively reducing oscillations throughout the robot’s motion and achieving an average energy saving of approximately 80 %. These results demonstrate that robust, adaptive control in autonomous two-wheeled robotic systems can be achieved by fusing data-driven models with classical control techniques.

Çetin, Gökhan; Ünker, Faruk

A two-wheeled robot has high maneuverability and can spin around with a high velocity but is prone to roll over under a recoil force. Because the wheels are only used to control the maneuver speed of the robot. However, a two-wheeled robot using a gyroscope does not need the control of the wheels and can maintain the stability of the robot under a recoil force during maneuvers even if the gimbal of the gyroscope is not controlled by a torque. In this study, the equations of motion of the gyroscopic robot are derived via Lagrangian equations, and the balancing performance of the uncontrolled gyroscope is simulated via Matlab to determine a desired configuration for mechatronics system design. Then, a linear-quadratic regulator (LQR) is implemented to the chosen control moment gyroscope (CMG) configuration and compared in simulations of RecurDyn. LQR controller was used to prevent oscillating motions of double CMGs under the recoil force of a weapon. The simulation results showed that the gyroscopic two-wheeled robot is improved using an LQR controller and powerfully balanced in static with excellent performance.

Çetin, Gökhan; Ünker, Faruk

The control moment gyroscope’s (CMG) success in emerging damping control under impact load is of great interest for new technical applications. This paper explores the potential use of a two-wheeled robot with an independent weapon, such as an automatic cannon robot, and considers employing CMG with LQR control as a weapon recoil stabiliser for the robot. Cannon turret firing accuracy is crucial for optimising LQR control parameters and enhancing weapon-robot performance. CMG with LQR control effectively reduces the recoil impact loads of the overhead weapon while compensating angular momentum to ensure safety for robot in firing. The precession motion of the gimbal from several milliradians to tens of milliradians is one of some contradictions for describing the effect of the recoil force on robot–gyroscope performance. This paper uses computational methods, combined with CAE software, to study the impact of increasing angular momentum on recoil-induced weapon vibrations.

Ünker, Faruk

In this study, an angular momentum generated by a flywheel gyroscope is performed as a suspension to prevent vehicle vibration. The frequency equations are derived and the optimal flywheel speed is obtained for vibration attenuation of the sprung mass. It is observed that the control moment gyroscope (CMG) keeps the sprung mass in the upright position with a small vibration amplitude of a steady motion. The flywheel speed (Ω) is in relation to the amplitude of gimbal motion and the frequency (ω) of the axle’s excitation. Besides, a simulation model of CAE software (RecurDyn) is built to validate the equations of motion.

Ünker, Faruk

Two-wheeled robots are popular in transportation applications because of their high maneuverability. In this research, the oscillation attenuation performance of the control moment gyroscope (CMG) for the two-wheeled robot was studied. The stored kinetic energy of a CMG can offer a weight and volume saving compared to conventional vibration absorbers. This CMG is also more reactionless than other conventional absorbers by transforming the impact of angular momentum to unidirectional thrust along the center of gravity. The gimbals can precess while providing the angular momentum under the gravitational force. This study indicated that the CMG can operate in a wide range of excitation frequencies to balance the robot in a stable period. Because the flywheel speed is much easier changed to thrust against the unwanted oscillations disturbing the robot stability. There is a relation between the gimbal amplitude and the flywheel speed of CMG, in which the required flywheel speed can be reduced if the higher gimbal amplitude is chosen. It can be also concluded from the study that the oscillation amplitudes at the target frequency can decrease as much as flywheel speed increases. There was also a mathematical model using ANSYS software. The simulation results using ANSYS matched well with the theoretical results of the Lagrangian model.

Ünker, Faruk; Cuvalci, Olkan

In this study, a gyroscopic damper model consisting of a built-in column has been developed. The objective of this study is determining the required angular momentum with the optimum ratio of natural frequency for the gyroscopic vibration absorber under sinusoidal excitation. This damper model is investigated experimentally and theoretically for different angular momentums of the gyroscope and different frequency ratios of the structure and damper on a single-story structure. As a result of the study, the displacement of the structure at the 1st mode frequency for a given angular momentum of the gyroscope is fairly reduced. Besides, at a notable frequency ratio of the damper and the structure, the effect of the angular momentum is improved. Even if the disk mass of the gyroscope decreases, it is possible to obtain a certain angular momentum by increasing the disk speed. Thus, lighter and less bulky gyroscopic vibration dampers can be designed. The cooperation of the required angular momentum with the optimum ratio of natural frequencies should be considered to obtain a highest attenuation of the building vibration. In addition, the equations of motion obtained according to the 1st mode and the theoretical results overlapped.

Ünker, Faruk

In this article, gyroscopes and flywheels are used to prevent the rollover of a vehicle due to external forces. The rollover preventing performance of the flywheels for a heavy trailer (an inverted pendulum problem) is investigated at a high road bank angle risk. Two control moment gyroscopes (CMG) and a reaction wheel are controlled by proportional torques to keep the vehicle in a stable motion in the vertical position. The reaction wheel was used only to eliminate the dissipation energy of damper. By using the energy stored in the flywheels of gyroscopes, the sprung mass of a vehicle can make a stable oscillating motion of small amplitude, standing upright without tipping over. The optimum flywheel speed was derived from the frequency equations with the appropriate controller gain. Most importantly, the required angular momentum to keep the vehicle upright without tipping over depends on the amplitude of the gimbal oscillation and the frequency. It has also been observed that Matlab simulations of Lagrangian equations and simulations modeled in RecurDyn software are in perfect harmony.

Ünker, Faruk

A two-wheeled self-balancing robot is considered for investigating the responses of a control moment gyroscope powered by a proportional controller to prevent the robot rollover against the constant inertia forces because of accelerations of the wheels of the robot. The amplitudes of the frequency equations related to the required angular momentum of flywheels with an optimum controller gain were also found. A simulation model of the robot using computer-aided engineering software (RecurDyn) is built to verify the equations of a Lagrangian model. The results of both obtained from the Lagrangian and that from RecurDyn simulations are analyzed comparatively, in which the proportional control loop reduces the required flywheel speeds Ω of gyros and keeps the robot in a very small amplitude of a stable sinusoidal motion in the upright position.

Çuvalci, Olkan; Ünker, Faruk; Baturalp, Turgut Batuhan; Gülbulak, Utku; Ertas, Atila

In this paper, an experimental model of a horizontal cantilever beam with a rotating/oscillating attached to the shaker for harmonic excitation at the one end and a gyrostabilizer at the other end is built to verify the equations of the Lagrangian model. The primary focus of the study was to investigate the parameters of excitation amplitude, natural frequency, rotating mass (disk mass), and disk speed of gyro that would minimize the amplitude of the beam to identify these effects. Numerical and experimental results indicate that the angular momentum of the gyrostabilizer is the most effective parameter in the reduction of beam displacement.

Ünker, Faruk

A special combined gyro-pendulum stabilizer (a gyroscope with coupling to a pendulum) mounted on a vibrating mass is considered for investigation of the vibration responses. This paper mainly focuses on the derivation of the frequency equations and on finding the required angular momentum for vibration control of the system. Besides, there is also an ANSYS simulation model of gyro-pendulum, which was built to verify the mathematical model. The dynamic responses of both that obtained from ANSYS simulation and that obtained from numerical solving of a Lagrangian mathematical model are analyzed comparatively. The angular momentum (ΩIp), in relation to the natural frequency (ωn) of the primary mass, shows that this vibration control device is more adaptable than other conventional ones by producing unidirectional thrust along the forcing excitation axis whilst the gyroscope is spinning.

Kaplan, Erdar; Ünker, Faruk; Çuvalcı, Olkan

-