Artificial intelligence control learning for autonomous industrial robots

Seismic events provide significant hazards to the safety and structural integrity of building structures, requiring efficient mitigation techniques. Conventional approaches to mitigating earthquake risks may lack the capacity for prompt adaptation and response. The integration of artificial intelligence (AI) controllers offers an attractive way to improve building safety in earthquake-prone regions. The fundamental unpredictability and complexity of seismic forces can undermine conventional safety measures, resulting in deficiencies in effectiveness and responses. A solution that can dynamically respond and proactively mitigate hazards must be developed to solve these concerns. This review attempts to analyze the possible impact of AI controllers in substantially mitigating seismic risks for structures. This study investigates the efficacy of intelligence structural security systems in enhancing resilience and reducing damage during seismic events through the analysis of AI-driven techniques, methodology, applications, and performance metrics. A systematic analysis of the literature is performed to identify and evaluate prior research on AI controllers employed to reduce seismic risk in structures. The research highlights the influence of integrated AI controllers on control systems, examining several AI controllers, including machine learning algorithms, neural networks, and evolutionary algorithms concerning structural safety. A case study is carried out on a conventional controller, specifically the sliding mode controller (SMC), fuzzy logic controller (FLC), and radial basis function neural network nonsingular terminal sliding mode controller (RBFNN-NTSMC). The findings reveal that the RBFNN-NTSMC effectively reduces building vibrations by up to 63% compared to an uncontrolled structure, significantly outperforming the FLC and SMC, which achieved reductions of 13% and 12%, respectively. This case study illustrates how AI-driven techniques enhance structural resilience and reduce seismic vulnerability. The integration of AI controllers has the potential to enhance the safety and durability of structures by employing advanced computational methods to limit hazards, facilitate real-time response, and optimize structural performance during earthquakes.

Обоснован перечень задач самообучения интеллектуальных систем управления автономными роботами и мультиагентными робототехническими группировками. Обсуждаются основные проблемы создания интеллектуальных самообучающихся систем управления. Показано, что один из наиболее перспективных подходов к реализации процессов самообучения в интеллектуальных системах управления автономными роботами и многоагентными робототехническими группировками связан с использованием метода автоматического построения деревьев классификации. Приводятся результаты комплексного моделирования.

Sharing a common context of perception is a prerequisite in order for several agents to develop a common understanding of a language. We propose a method, based on a simple imitative strategy, for transmitting a vocabulary from a teacher agent to a learner agent. A learner robot follows and thus implicitly imitates the movements of a teacher robot. While bounded by mutual following, learner and teacher agents are set in a position from which they share a common set of perceptions, that they can then correlate with an arbitrary set of signals, a vocabulary. The teacher robot leads the learner robot through a series of situations in which it teaches it a vocabulary to describe its observations. The learner robot grounds the teacher’s ‘words’ by associating them with its own perceptions. Learning is provided by a dynamical recurrent associative memory, an Artificial Neural Network architecture. The system is studied through simulations and physical experiments where the vocabulary concerns the agents’ movements and orientation. Experiments are successful, learning is fast and stable in the face of a significant amount of experimental noise. This work suggests then that a simple movement imitation strategy is an interesting scenario for the transmission of a language, as it is an easy means of getting the agents to share the same physical context.KeywordsMobile RobotRadio SignalCorrect WordTeacher AgentLearning ArchitectureThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Aiming at the problem that reactive power is too high caused by the control method of electric drive system of pump station, an artificial intelligence control method of reactive power of electric drive system of pump station is proposed. The transmission signal of the electric drive system of pumping station is processed by the non-linear operation of proportion, integral and differential through the structure of fuzzy PID algorithm. The transmission signal enters the nonlinear function estimator of the neural network controller, and through the function processing of the inverse process, the nonlinear mapping ability of the neural network is utilized to finally realize the artificial intelligence control. In order to verify the control reaction rate and the optimization effect of the proposed control method, compared with the traditional control method, the simulation comparison experiment and the actual comparison experiment were respectively designed. Experimental results: The simulation comparison experiments show that the control response rate of artificial intelligence control method is better than the traditional control method; the actual contrast experiment shows that the artificial intelligence method can effectively reduce the reactive power of the transmission system and can be popularized.

1. This research aims to study the edge computing devices with a faster processing and signal connecting function based on development of the physical layer and magnetic sensing devices. It is important matter for smart robot in multi-axis motor control. Based on the variable robot properties, it includes humanoid robots, wheeled machines, and fixed robots, respectively, are contented. This research is constructed as follows: to establishing an edge computing device system, the image vision system with neural-network learning theory to control robot arm path, multiple magnetic sensors in physical layer to analyze the signals of the motor, voltage, current, temperature, etc., are transmitted signal to the edge computing device and IoTs computing server system.2. IntroductionThe Industry 4.0 concept represents the trend of data exchange and automation in the manufacturing industry. It is defined by cloud computing, cyber-physical systems and the IoTs. In Industry 4.0, computers and automation technologies are integrated with robots and their components [1,2]. These robots and other components are remotely connected to a computer system, and the computer system is equipped with algorithms that can control and learning robots with minimal human intervention. For the motor device, the robot composes the control of the DC motor and relate to the accessories vital role. However, there are many factors that can misguide the controller's work, to make an unable to accurately control and/or have the desired dynamic behavior. For example, the resistance (Ohmic) of the coil winding of the motor is changed due to the temperature is different. If the vibration level is high and the wear of the rotor ball bearing becomes larger, the friction coefficient will also be affected. Therefore, the cloud computing environment of industrial factories, a continuous processing is required to re-evaluate the precise values of motor parameters and update the corresponding controller settings. In addition, it should be noted that the torque and speed sensors of the motor must have the correct selection to assist for achieving the correctness of the program.3. Theory and methods: multi-axis motor sensing and visual images of robotic armsCollaborative robots focus on intelligent motor control. Based on this construction, multiple sensor component signals are added. The physical information such as voltage, current, vibration, electric heating, magnetic flux sensing, and motor position is diagnosed through integrated control. This study, the artificial intelligent motor with the sensing signals, is used to control motor speed, positioning, constant horsepower, constant torque, force feedback, vibration resistance, overheating, overload, and instantaneous short circuit, as shown in Figure 1. The electromagnetic control of this paper is as follows:A. Single robotic arm function automation: Directly control the robot motor, at the same time, it has higher precision and speed to perform specific tasks repeatedly, but most of them cannot clearly actuate without an electromagnetic sensor.B. Part of the arm automation: It assists decision-making through environmental sensing input signals. In this study, robots use vision and sensors to determine the decision-making process, which is not easy to deal with accidental arm interference.C. Conditional autonomous robot: Control the arm according to the environmental monitoring behavior, but the external server can operate in real time.D. Highly autonomous robot: an environment that needs to be defined in an area or any situation without human operation.4. Experimental results and discussionBased on the motor control of intelligent robots, this research proposes a better and faster algorithm structure, which is a neural network-like combined with particle swarm method, which is used to calculate the convergence result of a lower loss function, which can accurately control the motor parameters and combine the IoTs, as shown in Figure 2. First, the robotic arm establishes multi-input sensing values, using the above algorithm to evaluate the robotic arm motor parameters, the prevention of interference from peripheral physical sensing, and the signal processing of external visual images. The issues raised based on past research experience include:(1) Establish edge and fog calculations to control the voltage and current of the motor, (2) establish neural network algorithm to calculate motor parameter, (3) construct the image vision system, (4) establish a smart robot platform, including wheel-shaped and human-shaped robots as experiments, which uses the above-mentioned deep learning theory to construct the test of the robot's motor motion and signal transmission response, (5) to use the multiple sensors, motor electromagnetic signals, piezoelectric sensors, and load signals, as the force feedback of the robot to calculate and judge. Finally, (6) to integrate the edge computing of the sensing layer, the fog computing of the signal processing layer, and faster computing layer through the IoTs cloud server. The IoTs chip is to use the dual-core Linkit-7697 (Taiwan’s MediaTek Inc). The chip can perform as a cloud computing and remote device, which indeed meet the requires of providing multi-sensing force feedback for cooperative robot handling objects. **

Inductive Logic Programming (ILP) induces first-order clausal theories from learning examples (positive and negative) and knowledge of the domain. Such theories can be used for gradually increasing the understanding of a robot about its world through iterative learning process. In this work we present a method for autonomous creation of negative examples for an ILP learner (i.e. sifting method). We also present a method for managing the iterative learning process (i.e. step transition method) for an autonomous robot that uses the ILP learner for learning. The sifting method uses `out-of-domain' values of the parameters involved in the learning process to create a set of possible negative examples. From these examples the robot autonomously selects those which allow it to efficiently learn better hypotheses. The step transition method enables the autonomous robot to decide how should it learn the new knowledge such that it can also gain profit from its experience. The robot makes this decision by comparing results of two different learning processes conducted on different data sets produced by the same action of the robot. The proposed methods are developed using the ILP learner Aleph, however they can also be used with other similar ILP learners. We experiment with these methods for learning primitive concepts for a mobile autonomous robot in a simple world. Results of the experiments show that the robot learns meaningful definitions of different physical notions in a hierarchical manner.

Purpose The following article is a “Q&A interview” conducted by Joanne Pransky of Industrial Robot Journal as a method to impart the combined technological, business, and personal experience of a prominent, robotic industry PhD and inventor regarding his pioneering efforts and the commercialization of bringing a technological invention to market. The paper aims to discuss these issues. Design/methodology/approach The interviewee is Dr Ken Goldberg, an inventor working at the intersection of art, robotics, and social media. He joined the UC Berkeley faculty in 1995 where he is the UC Berkeley William S. Floyd Jr Distinguished Chair in Engineering and recently served as Chair of the Industrial Engineering and Operations Research Department. He has secondary appointments in UC Berkeley’s Electrical Engineering/Computer Science, Art Practice and the School of Information. Goldberg also holds an appointment at the UC San Francisco Medical School’s Department of Radiation Oncology where he pursues research in medical robotics. Goldberg is Director of the CITRIS “People and Robots” Initiative and the UC Berkeley’s Laboratory for Automation Science and Engineering (AUTOLAB) where he and his students research machine learning for robotics and automation in warehouses, homes, and operating rooms. In this interview, Goldberg shares some of his personal and business perspectives from his career-long pursuit of making robots less clumsy. Findings Goldberg earned dual BS degrees in Electrical Engineering and Economics from the University of Pennsylvania in 1984, and MS and PhD degrees in Computer Science from Carnegie Mellon University in 1990. Goldberg also studied at Edinburgh University and the Technion. From 1991-95 he taught at the University of Southern California, and in fall 2000, he was visiting faculty at the MIT Media Lab. Goldberg and his students pursue research in three primary areas: Geometric Algorithms for Automation, Cloud Robotics, and Robot Learning. Originality/value Goldberg developed the first complete algorithms for part feeding and part fixturing, and developed the first robot on the Internet. His inventions have been awarded nine US Patents. Goldberg has published over 250 peer-reviewed technical papers and edited four books. He co-founded and served as Editor-in-Chief of the IEEE Transactions on Automation Science and Engineering (T-ASE). He is also Co-Founder of the Berkeley AI Research (BAIR) Lab, the Berkeley Center for New Media (BCNM), the African Robotics Network (AFRON), the Center for Automation and Learning for Medical Robotics (CAL-MR), the CITRIS Data and Democracy Initiative (DDI), Hybrid Wisdom Labs, and Moxie Institute. He has presented over four hundred keynote and invited lectures. Goldberg's artwork, closely linked with his research, has appeared in over seventy venues. Ken was awarded the Presidential Faculty Fellowship in 1995 by Bill Clinton, the Joseph Engelberger Robotics Award in 2000, elected IEEE Fellow in 2005, and selected by the IEEE Robotics and Automation Society for the George Saridis Leadership Award in 2016.

The relevance of choosing the subject of research as the basis for ensuring the safety of the use of robotic systems for various purposes, primarily those using artificial intelligence for control, and the object of research, which is the problem of sharing responsibility for the development and operation of robotic systems, is determined by the existing contradiction between the need for autonomous use of robotic systems and the complexity of software implementation of this requirement. At the same time, in robotics, quite often, it is the errors of control algorithms that are the source of most problems. Based on the analysis of regulatory documents regulating the development of artificial intelligence tools, possible problems of ensuring the safety of the use of autonomous robotic systems are analyzed. The conclusion is synthesized that, in the current state, these documents do not provide solutions to the security problem of artificial intelligence systems. A systematic approach was chosen as the methodological basis of the study. The use of a systematic approach, decomposition methods and comparative analysis made it possible to consider in a complex the problems of dividing the areas of responsibility of developers and operators of autonomous and partially autonomous robots implementing the principles of control based on artificial intelligence. The research's source base consists of scientific articles, regulatory and legislative documents that are publicly available. It is concluded that the existing approaches to training and self-learning of artificial intelligence systems that control autonomous robots “blur” the boundaries of responsibility of the participants in the process, which, in theory, can lead to critical situations during operation. With this in mind, based on the analysis of a typical development and application process, it is proposed to clarify the distribution of responsibility, as well as add new participants to the process: supplement it with specialists focused on the safety and impartiality of artificial intelligence (AI Alignment), and also provide a group approach in the development of artificial intelligence and machine learning algorithms, reducing the subjectivity factor.. Theoretically, the application of the principles of responsibility sharing synthesized in the article will ensure an increase in the safety of robotic systems based on the use of artificial intelligence.

We investigate the effects of behavioral activity on the development of patterns of synaptic connections in a model of neuromodulation embedded in an autonomous robot. Robot behavior produces clustered spatial distributions of rewarding objects in environments initially containing a high density of uniformly distributed objects. The spatial clustering of the rewarding objects restricts robot movements to sectors of high reward density, alters patterns of connections signaling predictions about reward timing, and changes the temporal profile of interactions between the robot and the objects. These effects are outcomes of embodied behavioral activity and are not pre-programmed or externally controlled. We discuss our results in the context of the reciprocal coupling of neural dynamics, behavioral activity and stimulus distributions.

This thesis presents the design and fabrication of an autonomous mobile robot with dual six-axis robotic arms, enhanced by augmented reality (AR) and powered by artificial intelligence (AI). The innovation aims to advance object manipulation tasks in various complex and dynamic environments, potentially for industrial applications. The system's Artificial Intelligence feature enhances the robot's ability to navigate and interact with its surroundings more accurately. The AR interface also allows human operators to intuitively control and guide the robot's actions in real time when needed, bridging the gap between autonomous operation and human oversight. The dual six-axis robotic arms enable dexterous and precise object manipulation. These arms mimic the versatility and adaptability of human arms, opening new possibilities for tasks that were previously challenging for traditional robotic systems. In developing this advanced mobile robot, this work embodies a pioneering step in autonomous robotics, showcasing a blend of AI, AR, and mechanical design. The robot's capabilities in object manipulation present a leap forward in robotics, promising a future where robots can take on more complex roles in our societies, while also offering a practical solution to labor-intensive or high-risk tasks. The research concludes with extensive testing and validation of the robot, demonstrating its effectiveness and efficiency in performing complex object manipulation tasks autonomously in various controlled and real-world scenarios. The results suggest promising implications for the future of robotics and automation, reinforcing the importance of interdisciplinary approaches in technological advancement.

In this study, adaptive artificial intelligence control algorithms were designed using the particle swarm optimisation (PSO) method for safe take-off and landing of aircraft. The performance of the designed control algorithms, which were applied to the control of the aircraft’s pitch angle, was compared with each other and with studies in the literature. Integrated squared error (ISE) was used as a fitness function in the design of PSO-based proportional–integral–derivative (PSO-PID), PSO-based fuzzy logic controller (PSO-FLC) and PSO-based self-adaptive fuzzy-PID (PSO-SAF-PID) controllers. Performance comparisons between the designed controllers were made based on the transient behaviours such as overshoot, rise time and settling time, and statistical error analyses including mean squared error (MSE), mean absolute error (MAE), ISE and integrated absolute error (IAE). It was observed that the performance of the PID controller, FLC and SAF-PID controllers improved significantly with the application of the PSO method. First, the mathematical model of the system was derived. Second, traditional control methods (PID control algorithm) and artificial intelligence control methods (FLC and SAF-PID type control algorithms) were designed to control the modelled system. Then, to improve the performance of these designed control algorithms, the optimal control parameters of the PID algorithm were determined using the PSO method. At the same time, membership function weight coefficients of the FLC algorithm and SAF-PID type control algorithm were determined using the PSO algorithm. Determining the control parameters and membership function coefficients at optimum values significantly affects the performance of control algorithms. Optimal control parameters and membership coefficients have significantly improved control performance. Furthermore, tests performed on transfer functions with different system parameters have shown that the PSO-SAF-PID controller is robust.

© 2016 IEEE. Uncertainties in base placements of mobile, autonomous industrial robots can cause incomplete coverage in tasks such as grit-blasting and spray painting. Sensing and localization errors can cause such uncertainties in robot base placements. This paper addresses the problem of collaborative complete coverage under uncertainties through appropriate base placements of multiple mobile and autonomous industrial robots while aiming to optimize the performance of the robot team. A mathematical model for complete coverage under uncertainties is proposed and then solved using a stochastic multi-objective optimization algorithm. The approach aims to concurrently find an optimal number and sequence of base placements for each robot such that the robot team's objectives are optimized whilst uncertainties are accounted for. Several case studies based on a real-world application using a realworld object and a complex simulated object are provided to demonstrate the effectiveness of the approach for different conditions and scenarios, e.g. various levels of uncertainties, different numbers of robots, and robots with different capabilities.

Estimation and valuation of confounding factors via artificial intelligent technologies is of the main importance when eliminating intraocular pressure tonometric mistakes. Current investigation focuses on intraocular pressure (IOP) magnitude, being the basis in diagnosis and monitoring of ocular hypertension, valuation reliability. The statistical modeling of experimental data, proved the central corneal thickness (CCT) to be confounding factor and positively correlated source of variation in intraocular pressure measurements among ocular hypertension subjects. To fix reliability of measured intraocular pressure magnitudes the artificial intelligent control in terms of numerical simulations via finite element method is proposed. Referring to statistical modeling of experimental data and provided numerical simulations in general conclusion we point the imperative of the artificial intelligent control method applied for introduce of CCT correction factor when having measured IOP via GAT. The convergence of proposed artificial intelligent control method proves it to be an appropriate alternative for ocular hypertension misdiagnose. Ill. 5, bibl. 3 (in English, summaries in Lithuanian, English, Russian).

A new artificial intelligent controller (AIC) is proposed for control of complex systems. This controller incorporates various approaches of artificial intelligence including central mechanism, fuzzy decision, and neural networks. Numerical simulations show the effectiveness of the proposed adaptive control design. >

This work proposes a machine-learning or artificial intelligence (AI) control of a low-drag Ahmed body with a rear slant angleφ = 35° with a view to finding strategies for efficient drag reduction (DR). The Reynolds numberReinvestigated is 1.7 × 105based on the square root of the body cross-sectional area. The control system comprises of five independently operated arrays of steady microjets blowing along the edges of the rear window and vertical base, twenty-six pressure taps on the rear end of the body and a controller based on an ant colony algorithm for unsupervised learning of a near-optimal control law. The cost function is designed such that bothDRand control power input are considered. The learning process of the AI control discovers forcing that produces aDRup to 18 %, corresponding to a drag coefficient reduction of 0.06, greatly exceeding any previously reportedDRfor this body. Furthermore, the discovered forcings may provide alternative solutions, i.e. a tremendously increased control efficiency given a small sacrifice inDR. Extensive flow measurements performed with and without control indicate significant alterations in the flow structure around the body, such as flow separation over the rear window, recirculation bubbles and C-pillar vortices, which are linked to the pressure rise on the window and base. The physical mechanism forDRis unveiled, along with a conceptual model for the altered flow structure under the optimum control or biggestDR. This mechanism is further compared with that under the highest control efficiency.