Recently, type-3 (T3) fuzzy logic systems (FLSs) have been widely used in various problems, such as modeling, control systems, image processing, forecasting problems, optimization algorithms, and many others. Most studies of T3-FLS focus on its different applications. However, the basic theory, the applications in real-time and online problems, learning schemes, and the robustness against non-Gaussian noises have been rarely studied. In this article, the simplification of T3-FLSs is taken into account, and the new membership functions (MFs), learning schemes, and type reduction are introduced. The concept of singleton MFs in adaptive fuzzy inference systems (ANFIS) is extended to T3-FLSs, and T3-ANFIS is proposed. The type reduction is simplified, and a noniterative learning scheme is developed. The corresponding computations for adaptation laws are derived, and all rules parameters and MF parameters are adjusted. To enhance the robustness versus impulsive noises, a T3-FLS-based correntropy Kalman filter (CKF) is designed. In the suggested algorithm, the kernel-size is not a constant value, but it is online updated by a T3-FLS. Also, to further improve robustness against noisy data, nonsingleton fuzzification for the suggested MF is formulated. By several simulations using real data sets, the feasibility of the suggested T3-FLS is shown, and its superiority is verified by comparisons. Also, the better robustness of suggested T3-FLS-based CKF versus impulsive noises is shown by comparison with traditional KFs.

This article introduces a robust AI-driven hybrid vision/force control (HVFC) method for industrial robots. The proposed HVFC method exploits Superpoint, a pretrained deep convolutional neural network (DCNN), as the AI agent to extract interest points for image-based visual servoing (IBVS), making it a target-object-free method. This tackles the limited workspace issue of eye-in-hand robots interacting with a workpiece due to the short distance between the camera and the workpiece, including a target object or landmarks. A learning-by-demonstration (LBD) method is also developed to generate the desired interest points associated with the desired path on the workpiece for interaction. To handle the issue of a high and variable number of interest points for use in IBVS, a set of six independent image features is extracted from the detected interest points, resulting in an invertible image interaction matrix, leading to global stability and a robust control process. To perform HVFC, a hierarchical orthogonal sliding manifold is defined, allowing force control in the normal direction and IBVS in the rest. Further, a filtered terminal integral sliding-mode controller is developed to stabilize the manifold, resulting in high tracking accuracy and robust performance against uncertainties and measurement noises. The experimental results of polishing and sanding the surfaces of a flat plastic board, a wooden airplane propeller, and a metal pegboard demonstrate the feasibility and superiority of the proposed HVFC-LBD method over conventional counterparts in terms of workspace expansion, robustness, and tracking accuracy.

Utilizing Uncrewed Aerial Vehicles (UAVs) offers a cost-effective and flexible option for various applications. However, achieving collision-free autonomous navigation requires advanced technology and safety assurances. This paper introduces a novel intelligent hybrid control scheme for UAV autonomous cruising and obstacle avoidance tasks. The new hybrid controller leverages deep reinforcement learning algorithms from our previous work with significant upgrades and incorporates a fuzzy logic model, greatly enhancing training efficiency. This paper presents the simulation results in 2D cases, which demonstrate the effectiveness of this approach. The results are also compared with the RL-only method presented in earlier works, highlighting the advantages of the new hybrid method. This research advances the field of safe autonomous navigation for UAVs under challenging airspace conditions.

The efficient control of serial robots in the presence of dynamic uncertainties and external disturbances is significant in many industrial applications. In this work, an adaptive sliding mode control (ASMC) approach with deep recurrent neural network (DRNN) is proposed for a 6-degree-of-freedom (6-DOF) industrial serial robot in joint space. A model-based sliding mode controller is developed to maintain the strong robustness of the robotic system. A deep recurrent neural network is designed to estimate the lumped system uncertainties in the controller. It consists of a feedforward structure through two hidden layers and a feedback loop from the output layer to the input layer, which exhibits more powerful online learning ability and dynamic property than shallow feedforward neural networks. According to Lyapunov theorem, the adaptation laws of the neural network parameters are derived, and the stability of the controller can be guaranteed. Simulation results demonstrate the effectiveness and superiority of the DRNN-based ASMC strategy regarding estimation convergence speed and trajectory tracking accuracy.

In this article, we present a novel full-pose path planning strategy with degenerate direction avoidance for robot manipulators. Effort seeks to define an efficient multiobjective cost function for finding an optimal path in task space by minimizing the path length, and, meanwhile, maximizing orthogonality to the degenerate direction of the path. Since a robot manipulator by nature is a noncommensurate system, the unit dual quaternion is used to unify the pose representation and formulate the cost function of path length. Additionally, to justify the robot’s ability to maneuver, another cost evaluation of orthogonality is developed on the basis of the analytical Jacobian matrix decomposition. The optimized rapidly exploring random tree (RRT*) is, then, applied to execute the synthesized cost function in planning a path for a six-degree-of-freedom (6-DOF) FANUC-M-20iA industrial robot. By comparing with two relevant path planning methods, simulation analysis results illustrate the superior performance of the proposed path planning approach in terms of path direction and joint change.

In this paper, a novel adaptive sliding mode control scheme with RBF (radial basis function) neural network-based tuning method is proposed for the trajectory tracking of a 6-RSS (Revolute-Spherical-Spherical) parallel robot in Cartesian space. Parallel robot is a highly nonlinear system with closed-chain mechanisms, which poses the major challenges to the controller design. The robust sliding mode controller is developed to deal with system uncertainties such as modeling errors, frictions, and disturbances. With strong adaptation and learning ability, RBF neural network is adopted to identify the parallel robot dynamics, and then the adaptive self-tuning of the control gains in the controller is realized, which is more flexible than manual tuning method and can guarantee the desired results of the changing system. The stability of the controller has been validated using Lyapunov theorem. Simulation results demonstrate that the proposed controller can achieve better tracking performance than the sliding mode controller with fixed control gains.

The accuracy and stability are two fundamental concerns of the visual servoing control system. This paper presents an enhanced image based visual servoing (IBVS) method which can increase the accuracy of visual servoing with guaranteed stability. The proposed controller combines PD control with sliding mode control (SMC) for a 6DOF manipulator. Specifically, the main benefits of this approach lie in simple structure and easy implementation due to PD control and good robustness to uncertainties due to SMC. Compared with conventional proportional or SMC controller, this approach owns faster convergence velocity, and better global disturbance rejection ability. The stability of the enhanced IBVS method is proved by using Lyapunov method. Simulation and experimental results show that the proposed controller can increase the accuracy and robustness of a 6DOF robotic system.

Many well-established singular perturbation theories for singularly perturbed systems require the full knowledge of system model parameters. In order to obtain an accurate and faithful model, a new identification scheme for singularly perturbed nonlinear system using multitime-scale recurrent high-order neural networks (NNs) is proposed in this paper. Inspired by the optimal bounded ellipsoid algorithm, which is originally designed for discrete-time systems, a novel weight updating law is developed for continuous-time NNs identification process. Compared with other widely used gradient-descent updating algorithms, this new method can achieve faster convergence, due to its adaptively adjusted learning rate. Based on the identification results, a control scheme using singular perturbation theories is developed. By using singular perturbation methods, the system order is reduced, and the controller structure is simplified. The closed-loop stability is analyzed and the convergence of system states is guaranteed. The effectiveness of the identification and the control scheme is demonstrated by simulation results.

Hyperspectral imaging provides new opportunities for improving face recognition accuracy. However, it poses such challenges as difficulty in data acquisition, low signal to noise ratio (SNR), and high dimensionality. In this paper, we propose a novel method for hyperspectral face recognition with good recognition rates. We first reduce noise adaptively from each spectral band and then crop each face according to eye coordinates. We perform minimum noise fraction (MNF) transform to the cropped face data cube in order to extract a number of MNF bands. We perform log-polar transform to each MNF band and extract Fourier spectra from them. In this way, the extracted features are translation-, rotation-, and scale-invariant. We conducted some experiments to test this new method for hyperspectral face recognition with very promising results. For Hong Kong Polytechnic University Hyperspectral Face Database (PolyU-HSPD), we achieved an average correct recognition rate of 92.70% with standard deviation of 2.61 (92.70%±2.61), which is better than 3D Gabor wavelet face recognition (91.30%±2.09), local binary pattern (LBP) classification (90%±2.42), spectral signature-based classification (24.60%±3.87), spectral angle-based classification (25.40%±4.36), and collaborative representation-based classification (CRC) (86.10%±2.37) for this database. When we use all spectral bands in the face dataset, we obtain a correct classification rate of 93.49%±2.53.

The problem of multi-objective feedback controller design of nonlinear systems is solved in this paper. The T-S fuzzy model is adopted to describe the nonlinear systems and genetic algorithm is used to identify the T-S fuzzy model. The identified T-S fuzzy model is reduced by applying Higher Order Singular Value Decomposition (HOSVD) method. Based on the reduced T-S fuzzy model, an optimal state feedback controller is designed by achieving the trade-off among three conflicting object functions using the optimal Pareto frontier. The simulation results reveal the effectiveness of the proposed method.

This paper deals with the adaptive nonlinear identification and trajectory tracking via dynamic multilayer neural network (NN) with different timescales. Two NN identifiers are proposed for nonlinear systems identification via dynamic NNs with different timescales including both fast and slow phenomenon. The first NN identifier uses the output signals from the actual system for the system identification. In the second NN identifier, all the output signals from nonlinear system are replaced with the state variables of the NNs. The online identification algorithms for both NN identifier parameters are proposed using Lyapunov function and singularly perturbed techniques. With the identified NN models, two indirect adaptive NN controllers for the nonlinear systems containing slow and fast dynamic processes are developed. For both developed adaptive NN controllers, the trajectory errors are analyzed and the stability of the systems is proved. Simulation results show that the controller based on the second identifier has better performance than that of the first identifier.

In this paper, robust invariance switching control problem for a class of multi-input uncertain nonlinear cascade systems with structural uncertainties is studied. By using methods of passivity and switching control of states of the linear subsystem, both stabilization of the linear subsystem and invariance of the specified region in state space can be guaranteed simultaneously. Hence, a sufficient condition for the robust invariance switching control of this class of systems is derived. Under some additional assumptions, the whole system is semi-global asymptotically stable (Semi-GAS). Input-state stability (ISS) assumptions or linear growth conditions on nonlinearities are not required. Simulation results demonstrate the effectiveness of the proposed design.

In this paper, a switching control of Image Based Visual Servoing (IBVS) with laser pointer is introduced to control the pose of the end effector with respect to the stationary object in a robotic assembly system so that the image features observed by the camera reach the desired image features as fast as possible. The simple off-the-shelf laser pointer is adopted to realize the depth estimation for obtaining the image Jacobian matrices. By using a laser spot as an image feature and the partition Degree-Of-Freedom (DOF) method, the proposed switching control algorithm decouples the rotational and translational motion control of the robotic end effector to avoid the inherent drawbacks of traditional IBVS, such as image singularities, image local minima and relative long trajectory in Cartesian space. The experimental results on a robotic assembly system are given to verify the effectiveness of the proposed method.

Nowadays, Minimally Invasive Surgery (MIS) has gained enormous popularity among the surgeons and in teleoperation procedures such as Robotic Assisted Minimally Invasive Surgery and Tele-Robotic Minimally Invasive Surgery. Despite its many advantages, MIS decreases the tactile sensory perception of the surgeon during grasping or manipulation of biological tissues. The loss of tactile perception has attracted a lot of attention. In this paper, we describe the detailed design of the hardware and software system used for graphical representation of the tactile sensing data. The proposed hardware and data acquisition system receives signals from the sensors incorporated on the MIS grasper and then transmits them to a personal computer. The developed software is also presented and its capabilities are discussed accordingly. Using this designed system, it is possible to determine the softness of the grasped objects. The degree of the softness is also displayed visually on the computer. The tactile sensor consists of four sensing elements which are integrated in each jaw of a modified commercial endoscopic grasper. Each sensing element consists of two piezoelectric polymer Plyvinylidene Fluoride (PVDF) films. The combination of the output voltages from each sensing element is used to determine the softness of the grasped object.

A contour-based feature extraction method is proposed by using the dual-tree complex wavelet transform and the Fourier transform. Features are extracted from the 1D signals r and θ, and hence the processing memory and time are reduced. The approximate shift-invariant property of the dual-tree complex wavelet transform and the Fourier transform guarantee that this method is invariant to translation, rotation and scaling. The method is used to recognize aircrafts from different rotation angles and scaling factors. Experimental results show that it achieves better recognition rates than that which uses only the Fourier features and Granlund's method. Its success is due to the desirable shift invariant property of the dual-tree complex wavelet transform, the translation invariant property of the Fourier spectrum, and our new complete representation of the outer contour of the pattern.