This paper presents navigation and control of a robot for capturing a moving target using the vision system. A stereo camera is used to calculate the pose of the moving target (position and orientation). An Adaptive Unscented Kalman Filter (AUKF) is used to generate an optimal path to capture the moving object by estimating the state vector (position, orientation, linear and angular velocities) of the target. The Fuzzy Logic Adaptive System (FLAS) has been used to prevent the AUKF from divergence. The FLAS can evaluate the performance of UKF and tuning the factors in the weighted covariance to improve the accuracy of UKF. A new trajectory planning method for the space robot is proposed based on the information acquired from the vision system and estimation the linear and angular velocities of the target by AUKF. The results from simulation experiments were presented and discussed. It was concluded that the Fuzzy Adaptive Unscented Kalman Filter methods give more accurate results rather than the Unscented Kalman filter or Extended Kalman Filter.
This paper presents a new methodology to guide and control flexible wings UAV based on a stereo vision system with advanced fuzzy logic algorithms. This system allows to detect wing's deflections and shapes by using the novel Deflection-Detection Vision System (DDVS). This methodology has three different steps. First, the shape of the wing was identified by the determinant the deflection points on the wing using the stereo camera. Second, the fuzzy logic algorithm was used to classify the shapes of the wing and to determine the flight parameters such as, speed, angle of attack and roll angle. Finally, an autopilot controller based on intelligent, Adaptive Neuro-Fuzzy Inference System (ANFIS) algorithm was designed. Extensive, experimental studies were performed in wind tunnel to determine the characteristics of the flexible wing. The testing were performed for wind speeds ranging from 11 to 31 km/h, angles of attack ranging from -20 deg to +20 deg. The experimental results were extended and verified with simulation experiments for wider ranges of velocities and angle of attack.
Quadrotor UAVs present a significant control challenge due to their underactuated nature; strong coupling effects; nonlinear dynamics; and high sensitivity to unknown effect parameters, external disturbances, and uncertainties. To address this issue, this study proposes a Lyapunov-based backstepping (LYP) controller that ensures robust stability and precise trajectory tracking. The controller employs an inner- and outer-loop architecture for coupled position and attitude control. Its performance is compared with Proportional–Integral–Derivative (PID) and Fractional-Order PID (FOPID) controllers under three scenarios: nominal conditions, external disturbances, and model parameter uncertainties. All controller gains are optimized using Particle Swarm Optimization (PSO). Simulation results, which are evaluated using time-domain metrics and root mean square error (RMSE), demonstrate that the proposed LYP controller achieves superior robustness, faster disturbance rejection, and improved tracking accuracy compared to both PID and FOPID controllers.