Brain-machine interfaces (BMIs) aim at building a direct communication pathway between the brain and an external device, and represent an area of research where significant progress has been made during the past decade. Based on BMIs, mind information can be read out by neural signals to control machine actuators, and machine-coded sensory information can be delivered to specific areas in the brain. BMI techniques provide the opportunity to integrate machine intelligence with biological intelligence at multiple levels to develop a more powerful intelligent system, thereby creating a new field called ‘cyborg intelligence’.       Traditional artificial intelligence (AI) involves learning how to emulate human-like intelligence and the creation of machines or software systems that exhibit intelligence. Successful AI includes natural language processing, speech recognition, smart search engines, face recognition, and Q&A systems. Although AI techniques show advantages in high-performance computation, probabilistic models, statistical reasoning, optimization, and almost unlimited storage, modern AI systems are unlikely to match humans in learning, high-level reasoning, and flexible adaptation to varying environments, which are recognized as the best features of biological intelligence. Because of their obvious complementary strengths, studies of the convergence of machine and biological intelligence, i.e., cyborg intelligence, are of great significance in maximizing their capabilities through their integration.       The central problems of cyborg intelligence include information fusion and representation in sensory-motor integration, cognitive computational models in brain-machine collaborations, statistical models for decoding and encoding brain signals, computational models and architecture for cyborg intelligence, and related data and computation standards. To explore this exciting new field, two workshops were held in Changsha and Hangzhou, China, in October and December 2013, respectively. More than 30 experts from the neuroscience, neuroengineering, and computer science fields came together to discuss scientific problems and research trends in the cyborg intelligence area. This special issue includes five of the most interesting and significant presentations describing state-of-the-art research in this field.

Eye-specific segregation in the dorsal lateral geniculate nucleus (dLGN) and superior colliculus (SC) starts from the embryonic stage and continues to develop postnatally until eye-opening in mice. However, there have been few systematic studies on the details of this developmental process. Here, we carried out time-dependent studies of eye-specific segregation in the dLGN and SC. Our results demonstrated that the development of eye-specific segregation in the SC is completed before postnatal day 12 (P12), which is earlier than in the dLGN (P20). During the whole period, ipsilateral and overlapping axonal projections decreased continuously in both the dLGN and SC. On the other hand, contralateral axonal projections showed little change, except for a slight decrease between P8 and P20 in the dLGN.

A lack of myoelectric sources after limb amputation is a critical challenge in the control of multifunctional motorized prostheses. To reconstruct myoelectric sources physiologically related to lost limbs, a newly proposed neural-function construction method, targeted muscle reinnervation (TMR), appears promising. Recent advances in the TMR technique suggest that TMR could provide additional motor command information for the control of multifunctional myoelectric prostheses. However, little is known about the nature of the physiological functional recovery of the reinnervated muscles. More understanding of the underlying mechanism of TMR could help us fine tune the technique to maximize its capability to achieve a much higher performance in the control of multifunctional prostheses. In this study, rats were used as an animal model for TMR surgery involving transferring a median nerve into the pectoralis major, which served as the target muscle. Intramuscular myoelectric signals reconstructed following TMR were recorded by implanted wire electrodes and analyzed to explore the nature of the neural-function reconstruction achieved by reinnervation of targeted muscles. Our results showed that the active myoelectric signal reconstructed in the targeted muscle was acquired one week after TMR surgery, and its amplitude gradually became stronger over time. These preliminary results from rats may serve as a basis for exploring the mechanism of neural-function reconstruction by the TMR technique in human subjects.

To listen to brain activity as a piece of music, we proposed the scale-free brainwave music (SFBM) technology, which could translate the scalp electroencephalogram (EEG) into music notes according to the power law of both EEG and music. In the current study, this methodology was further extended to a musical ensemble of two channels. First, EEG data from two selected channels are translated into musical instrument digital interface (MIDI) sequences, where the EEG parameters modulate the pitch, duration, and volume of each musical note. The phase synchronization index of the two channels is computed by a Hilbert transform. Then the two MIDI sequences are integrated into a chorus according to the phase synchronization index. The EEG with a high synchronization index is represented by more consonant musical intervals, while the low index is expressed by inconsonant musical intervals. The brain ensemble derived from real EEG segments illustrates differences in harmony and pitch distribution during the eyes-closed and eyes-open states. Furthermore, the scale-free phenomena exist in the brainwave ensemble. Therefore, the scale-free brain ensemble modulated by phase synchronization is a new attempt to express the EEG through an auditory and musical way, and it can be used for EEG monitoring and bio-feedback.

An intelligent wheelchair is devised, which is controlled by a coordinated mechanism based on a brain-computer interface (BCI) and speech recognition. By performing appropriate activities, users can navigate the wheelchair with four steering behaviors (start, stop, turn left, and turn right). Five healthy subjects participated in an indoor experiment. The results demonstrate the efficiency of the coordinated control mechanism with satisfactory path and time optimality ratios, and show that speech recognition is a fast and accurate supplement for BCI-based control systems. The proposed intelligent wheelchair is especially suitable for patients suffering from paralysis (especially those with aphasia) who can learn to pronounce only a single sound (e.g., ‘ah’).

Brain-computer interfaces (BCIs) can provide direct bidirectional communication between the brain and a machine. Recently, the BCI technique has been used in seizure control. Usually, a closed-loop system based on BCI is set up which delivers a therapic electrical stimulus only in response to seizure onsets. In this way, the side effects of neurostimulation can be greatly reduced. In this paper, a new BCI-based responsive stimulation system is proposed. With an efficient morphology-based seizure detector, seizure events can be identified in the early stages which trigger electrical stimulations to be sent to the cortex of the brain. The proposed system was tested on rats with penicillin-induced epileptic seizures. Online experiments show that 83% of the seizures could be detected successfully with a short average time delay of 3.11 s. With the therapy of the BCI-based seizure control system, most seizures were suppressed within 10 s. Compared with the control group, the average seizure duration was reduced by 30.7%. Therefore, the proposed system can control epileptic seizures effectively and has potential in clinical applications.

The suspension system is a key element in motor vehicles. Advancements in electronics and microprocessor technology have led to the realization of mechatronic suspensions. Since its introduction in some production motorcars in the 1980s, it has remained an area which sees active research and development, and this will likely continue for many years to come. With the aim of identifying current trends and future focus areas, this paper presents a review on the state-of-the-art of mechatronic suspensions. First, some commonly used classifications of mechatronic suspensions are presented. This is followed by a discussion on some of the actuating mechanisms used to provide control action. A survey is then reported on the many types of control approaches, including look-ahead preview, predictive, fuzzy logic, proportional–integral–derivative (PID), optimal, robust, adaptive, robust adaptive, and switching control. In conclusion, hydraulic actuators are most commonly used, but they impose high power requirements, limiting practical realizations of active suspensions. Electromagnetic actuators are seen to hold the promise of lower power requirements, and rigorous research and development should be conducted to make them commercially usable. Current focus on control methods that are robust to suspension parameter variations also seems to produce limited performance improvements, and future control approaches should be adaptive to the changeable driving conditions.

This paper first introduces the concept of a geogram that captures richer features to represent the objects. The spatiogram contains some moments upon the coordinates of the pixels corresponding to each bin, while the geogram contains information about the perimeter of grouped regions in addition to features in the spatiogram. Then we consider that a convergence process of mean shift is divided into obvious dynamic and steady states, and introduce a hybrid technique of feature description, to control the convergence process. Also, we propose a spline resampling to control the balance between computational cost and accuracy of particle filtering. Finally, we propose a boosting-refining approach, which is boosting the particles positioned in the ill-posed condition instead of eliminating the ill-posed particles, to refine the particles. It enables the estimation of the object state to obtain high accuracy. Experimental results show that our approach has promising discriminative capability in comparison with the state-of-the-art approaches.

In this study, we developed and tested a high-precision motion trajectory tracking controller of a pneumatic cylinder driven by four costless on/off solenoid valves rather than by a proportional directional control valve. The relationship between the pulse width modulation (PWM) of a signal’s duty cycle and control law was determined experimentally, and a mathematical model of the whole system established. Owing to unknown disturbances and unmodeled dynamics, there are considerable uncertain nonlinearities and parametric uncertainties in this pneumatic system. A modified direct adaptive robust controller (DARC) was constructed to cope with these issues. The controller employs a gradient type adaptation law based on discontinuous projection mapping to guarantee that estimated unknown model parameters stay within a known bounded region, and uses a deterministic robust control strategy to weaken the effects of unmodeled dynamics, disturbances, and parameter estimation errors. By using discontinuous projection mapping, the parameter adaptation law and the robust control law can be synthesized separately. A recursive backstepping technology is applied to account for unmatched model uncertainties. Kalman filters were designed separately to estimate the motion states and the derivative of the intermediate control law in synthesizing the deterministic robust control law. Experimental results illustrate the effectiveness of the proposed controller.

For the proportional directional valve controlled by two proportional solenoids, the normal control method (NCM) energizes only one solenoid at a time. The performance of the valve is greatly influenced by the nonlinearity of the proportional solenoid, such as dead zone and low force gain with a small current, and this effect cannot be eliminated by a simple dead-zone current compensation. To avoid this disadvantage, we propose the differential control method (DCM). By employing DCM, the controller outputs differential signals to simultaneously energize both solenoids of the proportional valve, and the operating point is found by analyzing the force output of the two solenoids to make a minimum variation of the current force gain. The comparisons of the valve response characteristics are made between NCM and DCM by nonlinear dynamic simulation and experiments. Simulation and experimental results show that by using DCM, the frequency response of the valve is greatly enhanced, especially when the input is small, which means that the dynamic characteristics of the proportional valve are improved.

The shapes of speakers’ vocal organs change under their different emotional states, which leads to the deviation of the emotional acoustic space of short-time features from the neutral acoustic space and thereby the degradation of the speaker recognition performance. Features deviating greatly from the neutral acoustic space are considered as mismatched features, and they negatively affect speaker recognition systems. Emotion variation produces different feature deformations for different phonemes, so it is reasonable to build a finer model to detect mismatched features under each phoneme. However, given the difficulty of phoneme recognition, three sorts of acoustic class recognition—phoneme classes, Gaussian mixture model (GMM) tokenizer, and probabilistic GMM tokenizer—are proposed to replace phoneme recognition. We propose feature pruning and feature regulation methods to process the mismatched features to improve speaker recognition performance. As for the feature regulation method, a strategy of maximizing the between-class distance and minimizing the within-class distance is adopted to train the transformation matrix to regulate the mismatched features. Experiments conducted on the Mandarin affective speech corpus (MASC) show that our feature pruning and feature regulation methods increase the identification rate (IR) by 3.64% and 6.77%, compared with the baseline GMM-UBM (universal background model) algorithm. Also, corresponding IR increases of 2.09% and 3.32% can be obtained with our methods when applied to the state-of-the-art algorithm i-vector.

In this study, a two-hop wireless sensor network with multiple relay nodes is considered where the amplify-and-forward (AF) scheme is employed. Two algorithms are presented to jointly consider interference suppression and power allocation (PA) based on the minimization of the symbol error rate (SER) criterion. A stochastic gradient (SG) algorithm is developed on the basis of the minimum-SER (MSER) criterion to jointly update the parameter vectors that allocate the power levels among the relay sensors subject to a total power constraint and the linear receiver. In addition, a conjugate gradient (CG) algorithm is developed on the basis of the SER criterion. A centralized algorithm is designed at the fusion center. Destination nodes transmit the quantized information of the PA vector to the relay nodes through a limited-feedback channel. The complexity and convergence analysis of the proposed algorithms are carried out. Simulation results show that the proposed two adaptive algorithms significantly outperform the other previously reported algorithms.

With the deployment of heterogeneous networks, mobile users are expecting ubiquitous connectivity when using applications. For bandwidth-intensive applications such as Internet Protocol Television (IPTV), multimedia contents are typically transmitted using a multicast delivery method due to its bandwidth efficiency. However, not all networks support multicasting. Multicasting alone could lead to service disruption when the users move from a multicast-capable network to a non-multicast network. In this paper, we propose a handover scheme called application layer seamless switching (ALSS) to provide smooth real-time multimedia delivery across unicast and multicast networks. ALSS adopts a soft handover to achieve seamless playback during the handover period. A real-time streaming testbed is implemented to investigate the overall handover performance, especially the overlapping period where both network interfaces are receiving audio and video packets. Both the quality of service (QoS) and objective-mapped quality of experience (QoE) metrics are measured. Experimental results show that the overlapping period takes a minimum of 56 and 4 ms for multicast-to-unicast (M2U) and unicast-to-multicast (U2M) handover, respectively. The measured peak signal-to-noise ratio (PSNR) confirms that the frame-by-frame quality of the streamed video during the handover is at least 33 dB, which is categorized as good based on ITU-T recommendations. The estimated mean opinion score (MOS) in terms of video playback smoothness is also at a satisfactory level.