Adaptive neural network control with optimal number of hidden nodes for trajectory tracking of robot manipulators

Trajectory Analysis of 6-DOF Industrial Robot Manipulators by Using Artificial Neural Networks

Robot manipulators are robotic systems that are frequently used in automation systems and able to provide increased speed, precision, and efficiency in the industrial applications. Due to their nonlinear and complex nature, it is crucial to optimize the robot manipulator systems in terms of trajectory control. In this study, positioning analyses based on artificial neural networks (ANNs) were performed for robot manipulator systems used in the textile industry, and the optimal ANN model for the high-accuracy positioning was improved. The inverse kinematic analyses of a 6-degree-of-freedom (DOF) industrial denim robot manipulator were carried out via four different learning algorithms, delta-bar-delta (DBD), online back propagation (OBP), quick back propagation (QBP), and random back propagation (RBP), for the proposed neural network predictor. From the results obtained, it was observed that the QBP-based 3-10-6 type ANN structure produced the optimal results in terms of estimation and modeling of trajectory control. In addition, the 3-5-6 type ANN structure was also improved, and its root mean square error (RMSE) and statistical R2 performances were compared with that of the 3-10-6 ANN structure. Consequently, it can be concluded that the proposed neural predictors can successfully be employed in real-time industrial applications for robot manipulator trajectory analysis.

A model-free terminal sliding mode control for robots: Achieving fixed-time prescribed performance and convergence

This paper introduces a new control strategy for robot manipulators, specifically designed to tackle the challenges associated with traditional model-based sliding mode (SM) controller design. These challenges include the need for accurately computed system models, knowledge of disturbance upper bounds, fixed-time convergence, prescribed performance, and the generation of chattering. To overcome these obstacles, we propose the incorporation of a neural network (NN) that effectively addresses these issues by removing the constraint of a precise system model. Additionally, we introduce a novel fixed-time prescribed performance control (PPC) to enhance response performance and position-tracking accuracy, while effectively limiting overshoot and maintaining steady-state error within the predefined range. To expedite the convergence of the SM surface to its equilibrium point, we introduce a faster terminal sliding mode (TSM) surface and a novel fixed-time reaching control algorithm (RCA) with adaptable factors. By integrating these approaches, we develop a novel control strategy that successfully achieves the desired goals for robot manipulators. The effectiveness and stability of the proposed approach are validated through extensive simulations on a 3-DOF SAMSUNG FARA-AT2 robot manipulator, utilizing both Lyapunov criteria and performance evaluations. The results demonstrate improved convergence rate and tracking accuracy, reduced chattering, and enhanced controller robustness.

CBMC: A Biomimetic Approach for Control of a 7-Degree of Freedom Robotic Arm

Many approaches inspired by brain science have been proposed for robotic control, specifically targeting situations where knowledge of the dynamic model is unavailable. This is crucial because dynamic model inaccuracies and variations can occur during the robot's operation. In this paper, inspired by the central nervous system (CNS), we present a CNS-based Biomimetic Motor Control (CBMC) approach consisting of four modules. The first module consists of a cerebellum-like spiking neural network that employs spiking timing-dependent plasticity to learn the dynamics mechanisms and adjust the synapses connecting the spiking neurons. The second module constructed using an artificial neural network, mimicking the regulation ability of the cerebral cortex to the cerebellum in the CNS, learns by reinforcement learning to supervise the cerebellum module with instructive input. The third and last modules are the cerebral sensory module and the spinal cord module, which deal with sensory input and provide modulation to torque commands, respectively. To validate our method, CBMC was applied to the trajectory tracking control of a 7-DoF robotic arm in simulation. Finally, experiments are conducted on the robotic arm using various payloads, and the results of these experiments clearly demonstrate the effectiveness of the proposed methodology.

UDE-based task space tracking control of uncertain robot manipulator with input saturation and output constraint

This paper investigates the trajectory tracking problem of uncertain robot manipulators with input saturation and output constraints. Uncertainty and disturbance estimator (UDE) is used to tackle the model uncertainties and external disturbances. Different from most existing methods, UDE only needs the bandwidth of the unknown plant model for design, which makes it easy to be implemented. Nonlinear state-dependent function is employed to cope with output constraints and a second order auxiliary system is constructed to solve the input saturation. Finally, an UDE-based tracking controller is proposed based on the backstepping method. With the proposed control scheme, the input saturation and the output constraints are not violated, and all signals in the closed-loop system are bounded. The comparative simulation results of a two-link robot manipulator are utilized to validate the effectiveness and superiority of the proposed control method.

Geographic Information System Technology Combined with Back Propagation Neural Network in Groundwater Quality Monitoring

This study was conducted to explore the distribution and changes of groundwater resources in the research area, and to promote the application of geographic information system (GIS) technology and its deep learning methods in chemical type distribution and water quality prediction of groundwater. The Shiyang River Basin in Minqin County was selected as the research object for analyzing the natural components distribution and its preliminary forecast in partial areas. With the priority control of groundwater pollutants, the concentration changes of four indicators (including the permanganate index) in different spatial distributions were analyzed based on the GIS technology, so as to provide a basis for the groundwater quality prediction. Taking the permanganate as a benchmark, this study evaluated the prediction effects of the conventional back propagation (BP) neural network (BPNN) model and the optimized BPNN based on the golden section (GBPNN) and wavelet transform (WBPNN). The algorithm proposed in this study is compared with several classic prediction algorithms for analysis. Groundwater quality level and distribution rules in the research area are evaluated with the proposed algorithm and GIS technology. The results reveal that GIS technology can characterize the spatial concentration distribution of natural indicators and analyze the chemical distribution of groundwater quality based on it. In contrast, the WBPNN has the best prediction result. Its average error of the whole process is 3.66%, and the errors corresponding to the six predicated values are all below 10%, which is dramatically better than the values of the other two models. The maximal prediction accuracy of the proposed algorithm is 97.68%, with an average accuracy of 96.12%. The prediction results on the water quality level are consistent with the actual condition, and the spatial distribution rules of the groundwater water quality can be shown clearly with the GIS technology combined with the proposed algorithm. Therefore, it is of great significance to explore the distribution and changes of regional groundwater quality, and this studywill play a critical role in determining the groundwater quality.

PD Control Compensation Based on a Cascade Neural Network Applied to a Robot Manipulator

A Proportional Integral Derivative (PID) controller is commonly used to carry out tasks like position tracking in the industrial robot manipulator controller; however, over time, the PID integral gain generates degradation within the controller, which then produces reduced stability and bandwidth. A proportional derivative (PD) controller has been proposed to deal with the increase in integral gain but is limited if gravity is not compensated for. In practice, the dynamic system non-linearities frequently are unknown or hard to obtain. Adaptive controllers are online schemes that are used to deal with systems that present non-linear and uncertainties dynamics. Adaptive controller use measured data of system trajectory in order to learn and compensate the uncertainties and external disturbances. However, these techniques can adopt more efficient learning methods in order to improve their performance. In this work, a nominal control law is used to achieve a sub-optimal performance, and a scheme based on a cascade neural network is implemented to act as a non-linear compensation whose task is to improve upon the performance of the nominal controller. The main contributions of this work are neural compensation based on a cascade neural networks and the function to update the weights of neural network used. The algorithm is implemented using radial basis function neural networks and a recompense function that leads longer traces for an identification problem. A two-degree-of-freedom robot manipulator is proposed to validate the proposed scheme and compare it with conventional PD control compensation.