Underactuated robot systems, due to their unique flexibility and economy, are widely used in modern industry and intelligent manufacturing. However, their underactuated nature and complex nonlinearity make the control problem challenging. In addition, most existing control methods for underactuated systems take the measurable actuated and unactuated states as an implicit premise and do not consider the unknown measurement sensitivity of sensors. Unfortunately, due to space/costs or manufacturing technology limitations in practical underactuated robot systems, it is difficult to measure unactuated states and achieve the ideal working mode of sensors. To this end, an adaptive fuzzy control scheme is proposed for a class of uncertain underactuated robot systems without unactuated state sensors, which stabilizes the system even with actuated state measurement errors. Specifically, the dynamics of underactuated systems are reconstructed into a nontriangular normal form, and the backstepping design is completed by using the boundedness of fuzzy basis functions. At the same time, the explosion of complexity is avoided by using dynamic surface control technology. Moreover, the proposed update law availably compensates parameter/structure uncertainties for underactuated robot systems. The asymptotic stability of the closed-loop system is proved by incorporating Lyapunov candidates with sensitivity information. Finally, the proposed control scheme is applied to a tower crane system, whose effectiveness is verified by hardware experiments.Note to Practitioners—This paper is motivated by the common sensor problems in uncertain underactuated robot systems control. Some states in underactuated systems are called unactuated states because they have no direct corresponding control inputs. Such states bring significant challenges to controllers design of underactuated systems. Furthermore, most existing control methods require all states to be measurable and ignore the measurement uncertainty of sensors. However, in many practical situations, it is difficult to install unactuated state sensors, and there are inevitable measurement errors in actuated state sensors. These factors make the feedback controllers developed for underactuated systems difficult to practically apply. To address these issues, this paper presents a control scheme that does not require unactuated state measurements and model information, which can still stabilize underactuated systems to the origin even in the presence of actuated state measurement errors. The stability is rigorously proven theoretically and the experimental results obtained on a self-built tower crane platform demonstrate the feasibility and effectiveness of the proposed control scheme. In future efforts, we intend to apply the proposed control scheme to practical industrial underactuated robot systems.

Underactuated systems are a class of systems in which the number of control inputs is less than the degrees of freedom (DoFs) to be controlled. With the increasing demand for the control performance of underactuated systems, the current research on their optimization of steady-state performance is no longer sufficient. However, owing to limited control inputs, ensuring their transient performance is often difficult. Moreover, some specific composite variables in underactuated systems should be kept within the preset ranges, which poses a significant challenge to collision avoidance safety. In addition, the sensor noises are also an issue that cannot be ignored. To this end, an extended Kalman filtering-based nonlinear model predictive control method for underactuated systems is developed in this article. The key feature of this method is that it simultaneously ensures accurate positioning, multiple constraints, and obstacle avoidance. Specifically, by adding an artificial potential field as an obstacle avoidance penalty term in the cost function and dynamically assigning weight coefficients, efficient collision avoidance control is achieved. Furthermore, it is combined with the extended Kalman filtering and jointly applied to underactuated systems with sensor noises. To the best of our knowledge, it is the first control method that simultaneously considers full-state constraints, specific composite variable constraints, control input and its increment constraints, as well as obstacle avoidance in underactuated systems. The satisfactory control performance of the proposed method is validated by implementing it on two typical underactuated systems, that is, four-DoF overhead cranes and five-DoF tower cranes.

Tower cranes play a vital role in various industries due to their extensive workspace and high lifting capacity. How-ever, their underactuated characteristics and complex nonlinear dynamics bring significant challenges for control. Previous studies have primarily focused on models and control strategies for point mass payload single pendulum cranes and point mass payload double pendulum cranes, overlooking practical considerations such as variable rope lengths and distributed mass payloads. To address the problems above, this paper uses Euler-Lagrange approaches to establish a dynamic model considering both distributed mass payloads and variable rope lengths. The proposed model avoids approximation and linearization to ensure accuracy and reliability. This research lays a theoretical foundation for advanced control strategies aimed at improving the safety, efficiency, and accuracy of tower crane operations.

Traffic flow prediction significantly affects the intelligent transportation for digitized urban transportation management and urban traffic control. Considering the complexity and strong non-linearity shown by traffic flow data, the establishment of model regarding spatial correlations as well as time dynamics can remarkably help to accurately predict traffic flow. A lot of current methods are mainly focused on using the historical time series information of observations to extract sequence features. Such forecasting will cause the lack of information and lead to poor accuracy of the forecast results. Although some studies applied spatial-temporal information, but they are not very accurate. In network-based problems, we would consider the constraint of road networks. Specifically, intersection flows, road speed and travel time are related to road networks. Also, they restrict the long-term prediction of traffic flow. For addressing above issues, a graph multi-head attention neural network (GMHANN) is proposed for the purpose of traffic flow prediction. In design, the GMHANN has an encoder-decoder structure. By the encoder, the data are compressed into a hidden space representation, which, relying on the decoder, is reconstructed as output. Furthermore, we put forward a novel gated recurrent unit (GRU) module (AGRU) based on multi-head attention for the effective extraction of the spatial and temporal features exhibited by traffic flow data. Other state-of-the-art methods are employed for evaluating four public datasets, which reveals that our proposed method outperforms others.

With the development of artificial intelligence, speech recognition and prediction have become one of the important research domains with wild applications, such as intelligent control, education, individual identification, and emotion analysis. Chinese poetry reading contains rich features of continuous pronunciations, such as mood, emotion, rhythm schemes, lyric reading, and artistic expression. Therefore, the prediction of the pronunciation characteristics of a Chinese poetry reading is the significance for the presentation of high-level machine intelligence and has the potential to create a high-level intelligent system for teaching children to read Tang poetry. Mel frequency cepstral coefficient (MFCC) is currently used to present important speech features. Due to the complexity and high degree of nonlinearity in poetry reading, however, there is a tough challenge facing accurate pronunciation feature prediction, that is, how to model complex spatial correlations and time dynamics, such as rhyme schemes. As for many current methods, they ignore the spatial and temporal characteristics in MFCC presentation. In addition, these methods are subjected to certain limitations on prediction for long-term performance. In order to solve these problems, we propose a novel spatial-temporal graph model (STGM-MHA) based on multihead attention for the purpose of pronunciation feature prediction of Chinese poetry. The STGM-MHA is designed using an encoder-decoder structure. The encoder compresses the data into a hidden space representation, while the decoder reconstructs the hidden space representation as output. In the model, a novel gated recurrent unit (GRU) module (AGRU) based on multihead attention is proposed to extract the spatial and temporal features of MFCC data effectively. The evaluation comparison of our proposed model versus state-of-the-art methods in six datasets reveals the clear advantage of the proposed model.

Passive compliance devices driven by cylinders have wide application prospects in industrial grinding areas with advantages of excellent passive adaptability and buffer capacity. However, there exists complex nonlinear hysteresis between the input pressure and the output force of cylinders. Moreover, the changes of the contact grinding force caused by various random external disturbances during the grinding process may severely reduce the grinding precision, and even cause irreversible dam-ages to the workpiece. Hence, how to keep the desired grinding force for compliance devices under the complex environment is an issue worth investigating. Focusing on the high precision grinding stage based on a compliance device driven by a double acting cylinder, the backlash-like hysteresis model is utilized to characterize the complex nonlinearities of the cylinder in this paper, and its inverse hysteresis is used to solve the desired pressure via elaborate iterative treatments. On this basis, an extended-state-observer-based nonlinear sliding mode controller is proposed. In addition, the closed-loop stability analysis is pro-vided via Lyapunov-based methods. Finally, the effectiveness and robustness of the proposed controller are verified by simulation results.

Many robots use range data to obtain an almost 3-dimensional description of their environment. Feature driven segmentation of range images has been primarily used for 3D object recognition, and hence the accuracy of the detected features is a prominent issue. Inspired by the structure and behaviour of the human visual system, we present an approach to feature extraction in range data using spiking neural networks and a biologically plausible hexagonal pixel arrangement. Standard digital images are converted into a hexagonal pixel representation and then processed using a spiking neural network with hexagonal shaped receptive fields; this approach is a step towards developing a robotic eye that closely mimics the human eye. The performance is compared with receptive fields implemented on standard rectangular images. Results illustrate that, using hexagonally shaped receptive fields, performance is improved over standard rectangular shaped receptive fields.

Inspired by the structure and behaviour of the human visual system, we present an approach to edge detection using spiking neural networks and a biologically plausible hexagonal pixel arrangement. Standard digital images are converted into a hexagonal pixel representation and then processed using a spiking neural network with hexagonal shaped receptive fields. The performance is compared with receptive fields implemented on standard rectangular images. Results illustrate that, using hexagonal shaped receptive fields, performance is improved over standard rectangular shaped receptive fields.

Spiking neural networks (SNNs) are an emerging computing paradigm that attempt to model the biological functions of the human brain. However, as networks approach the biological scale with significantly large numbers of neurons, software simulations face the problem of scalability and increasing computation times. Thus, numerous researchers have targeted hardware implementations in an attempt to more closely replicate the parallel processing capabilities of biological networks. Reconfigurable hardware is seen as a particularly viable platform for attempting to replicate to some degree the natural plasticity and flexibility of the human brain. This paper presents a scalable FPGA based implementation approach that facilitates the accelerated emulation of large-scale SNNs. The approach is validated using a SNN-based edge detection application where an order of magnitude speed performance increase was observed in comparison to a software equivalent implementation.

When an agent observes its environment, there are two important characteristics of the perceived information. One is the relevance of information and the other is redundancy. The irrelevant and redundant features which commonly exists within an environment, commonly leads to agent state explosion and associated high computational cost within the learning process. This paper presents an efficient method concerning both the relevance of information and the correlation in order to improve the learning of reinforcement learning agent. We introduce a new concurrent online learning method to calculate the match count C(s) and relevance degree I(s) to quantify the redundancy and correlation of features with respect to a desired learning task. Our analysis shows that the correlation relationship of the features can be extracted and projected to concurrent biased learning threads. By comparing the commonalities of these learning threads, we can evaluate the relevance degree of a feature that contributes to a particular learning task. We explain the method using random walk examples and then demonstrate the method on the chase object domain. Our validation results show that, using the concurrent learning method, we can efficiently detect redundancy and irrelevant features from the environment on sequential tasks, and significantly improve the efficiency of learning. After relevant features are extracted, the agent can remarkably accelerate its succeeding learning speed.

When symbolic AI approaches are applied to handle continuous valued attributes, there is a requirement to transform the continuous attribute values to symbolic data. In this paper, a novel distribution-index-based discretizer is proposed for such a transformation. Based on definitions of dichotomic entropy and a compound distributional index, a simple criterion is applied to discretize continuous attributes adaptively. The dichotomic entropy indicates the homogeneity degree of the decision value distribution, and is applied to determine the best splitting point. The compound distributional index combines both the homogeneity degrees of attribute value distributions and the decision value distribution, and is applied to determine which interval should be split further; thus, a potentially improved solution of the discretization problem can be found efficiently. Based on multiple reducts in rough set theory, a multiknowledge approach can attain high decision accuracy for information systems with a large number of attributes and missing values. In this paper, our discretizer is combined with the multiknowledge approach to further improve decision accuracy for information systems with continuous attributes. Experimental results on benchmark data sets show that the new discretizer can improve not only the multiknowledge approach, but also the naive Bayes classifier and the C5.0 tree

Knowledge representation is very important in intelligent systems -e.g. for knowledge discovery, data mining, and machine learning. The human brain, a significant intelligent system, works with a huge number of spiking neurons. Based on spiking neuron models a new generation of spiking neural networks (SNNs) has been developed for artificial intelligence systems. SNNs are computationally more powerful than conventional artificial neural networks. In this paper, the spiking neuron model is applied to represent logic rules and fuzzy rules. Based on the STDP (Spike Timing Dependent Plasticity) principle, a new SNN model is proposed for pattern recognition. An efficient learning rule derived from the STDP is applied for self-organizing the input training set efficiently. An example, Animal-Growth-Record, is used to explain the principle of the SNN model. Benchmark data sets are applied to compare the proposed approach with other approaches. As there are very efficient learning rules in the SNN model, the model can be applied not only for fusion of multi-sensory data, but also for data mining in large databases with large numbers of attributes.

A self-organizing computing network based on concepts of fuzzy conditions, beliefs, probabilities, and neural networks is proposed for decision-making in intelligent systems which are required to handle data sets with a diversity of data types. A sense-function with a sense-range and fuzzy edges is defined as a transfer function for connections from the input layer to the hidden layer in the network. By generating hidden cells and adjusting the parameters of the sense-functions, the network self-organizes and adapts to a training set. Computing cells in the input layer are designed as data converters so that the network can deal with both symbolic data and numeric data. Hidden computing cells in the network can be explained via fuzzy rules in a similar manner to those in fuzzy neural networks. The values in the output layer can be explained as a belief distribution over a decision space. The final decision is made by means of the winner-take-all rule. The approach was applied to a series of the benchmark data sets with a diversity of data types and comparative results obtained. Based on these results, the suitability of a range of data types for processing by different intelligent techniques was analyzed, and the results show that the proposed approach is better than other approaches for decision-making in information systems with mixed data types.

In this paper, an evolving learning mechanism is proposed for general computing network model to make decisions in intelligent systems. The novel mechanism is performed by means of a set of computing cell operations such as self-generation, growth, self-division, and death. Under the mechanism, a computing network grows up to a mature network. A hidden cell in the network is defined as a condition matching-unit in response to a fuzzy sub-superspace in multiple-dimension input superspace. A sense-function is defined to represent connections from a hidden cell to input cells. The range and edge vagueness of the sense-function are determined by evolving learning mechanism when sample instances are presented to the network. This network is able to learn from a very few training instances to make decisions for unseen instances. The benchmark data sets from the UCI machine learning repository are applied to test the network and comparable results are obtained.