Unactuated and Actuated States Simultaneously Constrained Optimal Trajectory Planning-Based Path-Following Control for Underactuated Robots.

For underactuated robots working in complex environments, an important objective is to drive all variables (particularly for unactuated end-effectors) to move along the specific path and restrict positions/velocities to avoid obstacles, rather than using only point-to-point control. Unfortunately, most path planning methods are only suitable to fully actuated systems or depend on linearized models. The main motivations of our work are to directly fulfill motion constraints and achieve path following for both actuated and unactuated states (e.g., payload swing of cranes) when lacking effective control inputs. To this end, this article presents a new time-optimal trajectory planning-based motion control method for general underactuated robots. By constructing auxiliary signals (in Cartesian space) to express all actuated/unactuated variables (in joint space), their position/velocity constraints are converted into some convex/nonconvex inequalities related to a to-be-optimized path parameter and its derivatives. Then, an optimization algorithm is constructed to solve the available path parameter and derive a group of time-optimal trajectories for actuated states. As we know, this is the first study to ensure path following and necessary full-state constraints for actuated/unactuated states. Then, a tradeoff among path-constrained motions, time optimization, and state constraints is achieved together. This article takes the rotary crane as an example and provides detailed analysis of calculating desired trajectories based on the proposed planning frame, whose effectiveness is also verified through hardware experiments.

Mind the Gap: Learning Modality-Agnostic Representations With a Cross-Modality UNet.

Cross-modality recognition has many important applications in science, law enforcement and entertainment. Popular methods to bridge the modality gap include reducing the distributional differences of representations of different modalities, learning indistinguishable representations or explicit modality transfer. The first two approaches suffer from the loss of discriminant information while removing the modality-specific variations. The third one heavily relies on the successful modality transfer, could face catastrophic performance drop when explicit modality transfers are not possible or difficult. To tackle this problem, we proposed a compact encoder-decoder neural module (cmUNet) to learn modality-agnostic representations while retaining identity-related information. This is achieved through cross-modality transformation and in-modality reconstruction, enhanced by an adversarial/perceptual loss which encourages indistinguishability of representations in the original sample space. For cross-modality matching, we propose MarrNet where cmUNet is connected to a standard feature extraction network which takes as inputs the modality-agnostic representations and outputs similarity scores for matching. We validated our method on five challenging tasks, namely Raman-infrared spectrum matching, cross-modality person re-identification and heterogeneous (photo-sketch, visible-near infrared and visible-thermal) face recognition, where MarrNet showed superior performance compared to state-of-the-art methods. Furthermore, it is observed that a cross-modality matching method could be biased to extract discriminant information from partial or even wrong regions, due to incompetence of dealing with modality gaps, which subsequently leads to poor generalization. We show that robustness to occlusions can be an indicator of whether a method can well bridge the modality gap. This, to our knowledge, has been largely neglected in the previous works. Our experiments demonstrated that MarrNet exhibited excellent robustness against disguises and occlusions, and outperformed existing methods with a large margin (>10%). The proposed cmUNet is a meta-approach and can be used as a building block for various applications.

Reconstructing Randomly Masked Spectra Helps DNNs Identify Discriminant Wavenumbers.

Nondestructive detection methods, based on vibrational spectroscopy, are vitally important in a wide range of applications including industrial chemistry, pharmacy and national defense. Recently, deep learning has been introduced into vibrational spectroscopy showing great potential. Different from images, text, etc. that offer large labeled data sets, vibrational spectroscopic data is very limited, which requires novel concepts beyond transfer and meta learning. To tackle this, we propose a task-enhanced augmentation network (TeaNet). The key component of TeaNet is a reconstruction module that inputs randomly masked spectra and outputs reconstructed samples that are similar to the original ones, but include additional variations learned from the domain. These augmented samples are used to train the classification model. The reconstruction and prediction parts are trained simultaneously, end-to-end with back-propagation. Results on both synthetic and real-world datasets verified the superiority of the proposed method. In the most difficult synthetic scenarios TeaNet outperformed CNN by 17%. We visualized and analysed the neuron responses of TeaNet and CNN, and found that TeaNet's ability to identify discriminant wavenumbers was excellent compared to CNN. Our approach is general and can be easily adapted to other domains, offering a solution to more accurate and interpretable few-shot learning.

Concurrent Learning-Based Adaptive Control of Underactuated Robotic Systems With Guaranteed Transient Performance for Both Actuated and Unactuated Motions.

With the wide applications of underactuated robotic systems, more complex tasks and higher safety demands are put forward. However, it is still an open issue to utilize "fewer" control inputs to satisfy control accuracy and transient performance with theoretical and practical guarantee, especially for unactuated variables. To this end, for underactuated robotic systems, this article designs an adaptive tracking controller to realize exponential convergence results, rather than only asymptotic stability or boundedness; meanwhile, unactuated states exponentially converge to a small enough bound, which is adjustable by control gains. The maximum motion ranges and convergence speed of all variables both exhibit satisfactory performance with higher safety and efficiency. Here, a data-driven concurrent learning (CL) method is proposed to compensate for unknown dynamics/disturbances and improve the estimate accuracy of parameters/weights, without the need for persistency of excitation or linear parametrization (LP) conditions. Then, a disturbance judgment mechanism is utilized to eliminate the detrimental impacts of external disturbances. As far as we know, for general underactuated systems with uncertainties/disturbances, it is the first time to theoretically and practically ensure transient performance and exponential convergence speed for unactuated states, and simultaneously obtain the exponential tracking result of actuated motions. Both theoretical analysis and hardware experiment results illustrate the effectiveness of the designed controller.

Neuroadaptive Control for Complicated Underactuated Systems With Simultaneous Output and Velocity Constraints Exerted on Both Actuated and Unactuated States.

Due to limited workspace and safety requirements for practical underactuated mechanical systems, it is necessary to restrict all to-be-controlled variables and their velocities within preset ranges, avoid collisions/overshoots, and improve braking performance. However, due to fewer available control inputs, it is quite challenging to ensure error elimination and full-state constraints for both actuated/unactuated variables, including displacements/angles and their derivatives (i.e., velocity signals) together. To handle the above issues, this article designs a new adaptive full-state constraint controller for a class of uncertain multi-input-multi-output (MIMO) underactuated systems. First, different output constraint-related auxiliary functions are constructed in the Lyapunov function candidate to generate nonlinear displacement-/angle-limited terms to control all state variables. Then, this article handles velocity constraints in a new manner, where the elaborately designed velocity constraint-related terms are directly introduced into the presented controller (instead of the Lyapunov function candidate), and strict theoretical analysis is provided by utilizing reduction to absurdity. Hence, both actuated and unactuated velocity constraints are ensured to further improve transient performance. In addition, the impact of model uncertainties is addressed online to realize accurate positioning control for all state variables. Compared with current studies of underactuated systems, this article presents the first adaptive controller to address output and velocity constraints for actuated and unactuated variables together; moreover, their asymptotic convergence is proven by strict stability analysis, which is important both theoretically and practically. In the end, the feasibility and robustness of the proposed controller are verified by hardware experiments.

Neural Network-Based Hybrid Three-Dimensional Position Control for a Flapping Wing Aerial Vehicle.

This article presents a novel neural network-based hybrid mode-switching control strategy, which successfully stabilizes the flapping wing aerial vehicle (FWAV) to the desired 3-D position. First, a novel description for the dynamics, resolved in the proposed vertical frame, is proposed to facilitate further position loop controller design. Then, a radial base function neural network (RBFNN)-based adaptive control strategy is proposed, which employs a switching strategy to keep the system away from dangerous flight conditions and achieve efficient flight. The learning process of the neural network pauses, resumes, or alternates its update strategy when switching between different modes. Moreover, saturation functions and barrier Lyapunov functions (BLFs) are introduced to constrain the lateral velocity within proper ranges. The closed-loop system is theoretically guaranteed to be semiglobally uniformly ultimately bounded with arbitrarily small bound, based on Lyapunov techniques and hybrid system analysis. Finally, experimental results demonstrate the excellent reliability and efficiency of the proposed controller. Compared to existing works, the innovations are the put forward of the vertical frame and the cooperative switching learning and control strategies.

A Reinforcement Learning-Based Strategy of Path Following for Snake Robots with an Onboard Camera.

For path following of snake robots, many model-based controllers have demonstrated strong tracking abilities. However, a satisfactory performance often relies on precise modelling and simplified assumptions. In addition, visual perception is also essential for autonomous closed-loop control, which renders the path following of snake robots even more challenging. Hence, a novel reinforcement learning-based hierarchical control framework is designed to enable a snake robot with an onboard camera to realize autonomous self-localization and path following. Specifically, firstly, a path following policy is trained in a hierarchical manner, in which the RL algorithm and gait knowledge are well combined. On this basis, the training efficiency is sufficiently optimized, and the path following performance of the control policy is greatly improved, which can then be implemented on a practical snake robot without any additional training. Subsequently, in order to promote visual self-localization during path following, a visual localization stabilization item is added to the reward function that trains the path following strategy, which endows a snake robot with smooth steering ability during locomotion, thereby guaranteeing the accuracy of visual localization and facilitating practical applications. Comparative simulations and experimental results are illustrated to exhibit the superior performance of the proposed hierarchical path following the control method in terms of convergence speed and tracking accuracy.

A Novel Learning-Based Trajectory Generation Strategy for a Quadrotor.

In this article, a learning-based trajectory generation framework is proposed for quadrotors, which guarantees real-time, efficient, and practice-reliable navigation by online making human-like decisions via reinforcement learning (RL) and imitation learning (IL). Specifically, inspired by human driving behavior and the perception range of sensors, a real-time local planner is designed by combining learning and optimization techniques, where the smooth and flexible trajectories are online planned efficiently in the observable area. In particular, the key problems in the framework, temporal optimality (time allocation), and spatial optimality (trajectory distribution) are solved by designing an RL policy, which provides human-like commands in real-time (e.g., slower or faster) to achieve better navigation, instead of generating traditional low-level motions. In this manner, real-time trajectories are calculated using convex optimization according to the efficient and accurate decisions of the RL policy. In addition, to improve generalization performance and to accelerate the training, an expert policy and IL are employed in the framework. Compared with existing works, the kernel contribution is to design a real-time practice-oriented intelligent trajectory generation framework for quadrotors, where human-like decision-making and model-based optimization are integrated to plan high-quality trajectories. The results of comparative experiments in known and unknown environments illustrate the superior performance of the proposed trajectory generation strategy in terms of efficiency, smoothness, and flexibility.

Low concentrations of the antidepressant venlafaxine affect courtship behaviour and alter serotonin and dopamine systems in zebrafish (Danio rerio).

Venlafaxine, a serotonin-noradrenaline reuptake inhibitor, is a widely used antidepressant drug routinely detected in aquatic environments. However, its potential impact on courtship behaviour in zebrafish is unknown. We tested the hypothesis that venlafaxine disrupts brain monoamine levels and molecular responses essential for courtship behaviour in zebrafish. Zebrafish (Danio rerio) were exposed to venlafaxine (1, 10, and 100 µg/L) for 20 days. We evaluated the molecular levels and neuronal basis of the effect of venlafaxine on courtship behaviour. Here, we show that venlafaxine inhibited courtship behaviour in zebrafish and increased the transcript levels of 5-ht1a and 5-ht2c while decreasing the transcript levels of genes involved in the dopaminergic system, including th1, th2, drd1b, and drd2b. Venlafaxine upregulated 5-HT levels and downregulated dopamine levels. Moreover, the subordinate fish from the venlafaxine-exposed group had significantly lower motor activity than the subordinate fish of the control group. Collectively, our results reveal that venlafaxine can disturb brain monoamine levels, affecting courtship behaviour in adult zebrafish.

Adaptive Fuzzy Control for a Class of MIMO Underactuated Systems With Plant Uncertainties and Actuator Deadzones: Design and Experiments.

In the field of modern industrial engineering, many mechanical systems are underactuated, exhibiting strong nonlinear characteristics and high flexibility. However, the lack of control inputs brings about many difficulties for controller design and stability/convergence analysis., some unavoidable practical issues, e.g., plant uncertainties and actuator deadzones, make the control of underactuated systems even more challenging. Hence, with the aid of elaborately constructed finite-time convergent surfaces, this article provides the first solution to address the control problem for a class of multi-input-multi-output (MIMO) underactuated systems subject to plant uncertainties and actuator deadzones. Specifically, this article overcomes the main obstacle in sliding-mode surface analysis for MIMO underactuated systems, that is, by the presented analysis method, the asymptotic stability of the system equilibrium point is strictly proven based on the composite surfaces. In addition, the unknown parts of the actuated/unactuated dynamic equations and actuator deadzones can be simultaneously handled, which is important for real applications. Furthermore, we apply the proposed method to two kinds of typical underactuated systems, that is: 1) tower cranes and 2) double-pendulum cranes, and implement a series of hardware experiments to verify its effectiveness and robustness.

An automated vertical drift correction algorithm for AFM images based on morphology prediction.

The atomic force microscope (AFM) has become a powerful tool in many fields. However, environmental noise and other disturbances are very likely to cause the AFM probe to vibrate, which lead to vertical drift in AFM imaging and limit its further application. Therefore, to correct image distortion caused by vertical drift, a morphology prediction based image correction algorithm is proposed in this paper. Specifically, a Gaussian-Hann filter is first designed for distorted AFM images, based on which, an adaptive image binarization algorithm is developed to achieve accurate object detection and background extraction. Furthermore, an advanced morphology prediction algorithm, consisting of morphological approximation prediction and morphological detail prediction, is proposed to correct image distortion by using the extracted substrate of a sample image. Approximate morphology is generated by an improved weighted fusion autoregressive model, and morphological detail is obtained by energy analysis based on discrete wavelet transform. Experimental and application results are presented to illustrate that the proposed algorithm is able to effectively eliminate vertical drift of AFM images.

Integrated navigation of GPS/INS based on fusion of recursive maximum likelihood IMM and Square-root Cubature Kalman filter.

Information fusion of the GPS/INS integrated system is always related to characteristics of the inertial system and the sensor feature, yet prior knowledge is still difficult to obtain in real applications. To deal with the uncertainty of error covariance and state noise in vehicle navigation, this paper presents a novel approach, wherein the integration of Square-root Cubature Kalman Filters (SCKF) and Interacting Multiple Model (IMM) are also introduced. In the framework of IMM, the SCKFs with different covariance are designed to reflect various vehicle dynamics. Besides, since the IMM-SCKF can switch flexibly among the filters, the transition probability matrix is computed with maximum likelihood method to adapt to different noise characteristics. The performance of the proposed algorithm is guaranteed by theoretical analyses, and a series of vehicular experiments with different maneuvers are carried out in an urban environment. The results indicate that, in comparison with the CKF and the IMM-CKF, the accuracy of velocity and attitude are increased by the proposed strategy.

A high-speed atomic force microscopy with super resolution based on path planning scanning.

An atomic force microscopy generally adopts a raster scanning method to obtain the image of the sample morphology. However, the raster method takes too much time on the base part without focusing enough on the object, thereby restricting the scanning speed of an AFM. To solve this problem, this paper proposes a novel path planning based scanning method to achieve high-speed scanning with super resolution for AFMs. Specifically speaking, a fast scanning process is first carried out to generate a low-resolution image with less time, then a convolutional neural network is designed to construct a super-resolution image based on the fast scanning image. Afterwards, an advanced detection algorithm is proposed to achieve the accurate object detection and localization. Furthermore, an improved ant colony optimization algorithm is proposed to realize the path planning for scanning the objects with high quality, whose imaging result is then matched with the previous super-resolution image to construct the entire sample image, thus achieving fast scanning with super resolution. Experimental and application results demonstrate the good performance of the proposed scanning method.

A Path-Integral-Based Reinforcement Learning Algorithm for Path Following of an Autoassembly Mobile Robot.

Reinforcement learning (RL) combined with deep neural networks has led to a number of great achievements for robot control in virtual computer environments, where sufficient data can be obtained without any difficulty to train various models. However, thus far, only few and relatively simple tasks have been accomplished for practical robots, which is mainly caused by the following two reasons. First, training with real robots, especially with dynamic systems, is too complicated to be fully and accurately represented in simulations. Second, it is very costly to obtain training data from real systems. To address these two problems effectively, in this article, a path-integral-based RL algorithm is proposed for the task of path following of an autoassembly mobile robot, wherein three kernel techniques are introduced. First, a generalized path-integral-control approach is proposed to obtain the numerical solution of a stochastic dynamical system, wherein the calculation of the gradient and kinematics inverse is avoided to ensure fast and reliable training convergence. Second, a novel parameterization method using Lyapunov techniques is introduced into the RL algorithm to ensure good performance of the system when directly transferring simulation results into practical systems. Third, the optimal parameters for all discrete initial states are first learned offline and then tuned online to improve the generalization and real-time performance. In addition to the optimization control for the mobile robot, the proposed method also possesses general applicability for a class of nonlinear systems such as crane systems. Simulation and experimental results are included and analyzed to illustrate the superior performance of the proposed algorithm.

Neural Network-Based Adaptive Antiswing Control of an Underactuated Ship-Mounted Crane With Roll Motions and Input Dead Zones.

As a type of indispensable oceanic transportation tools, ship-mounted crane systems are widely employed to transport cargoes and containers on vessels due to their extraordinary flexibility. However, various working requirements and the oceanic environment may cause some uncertain and unfavorable factors for ship-mounted crane control. In particular, to accomplish different control tasks, some plant parameters (e.g., boom lengths, payload masses, and so on) frequently change; hence, most existing model-based controllers cannot ensure satisfactory control performance any longer. For example, inaccurate gravity compensation may result in positioning errors. Additionally, due to ship roll motions caused by sea waves, residual payload swing generally exists, which may result in safety risks in practice. To solve the above-mentioned issues, this paper designs a neural network-based adaptive control method that can provide effective control for both actuated and unactuated state variables based on the original nonlinear ship-mounted crane dynamics without any linearizing operations. In particular, the proposed update law availably compensates parameter/structure uncertainties for ship-mounted crane systems. Based on a 2-D sliding surface, the boom and rope can arrive at their preset positions in finite time, and the payload swing can be completely suppressed. Furthermore, the problem of nonlinear input dead zones is also taken into account. The stability of the equilibrium point of all state variables in ship-mounted crane systems is theoretically proven by a rigorous Lyapunov-based analysis. The hardware experimental results verify the practicability and robustness of the presented control approach.

The effects of fluorene-9-bisphenol on female zebrafish (Danio rerio) reproductive and exploratory behaviors.

Endocrine disruptor chemicals induce adverse effects to animals' development, reproduction and behavior in environment. We investigated the effects of fluorene-9-bisphenol (BHPF), one substitute of bisphenol A, on courtship behavior and exploratory behavior of adult zebrafish. Customized apparatus was used to evaluate courtship behavior. The result showed that the male spent less time with BHPF and anti-oestrogenic fulvestrant (FULV) treated female in region of approaching (ROA). Courtship index between BHPF-exposed female and male decreased. The body orientation of BHPF- and FULV-exposed female to male decreased. Furthermore, BHPF exposure downregulated the expression of genes related to estrogen receptor, steroidogenesis and upregulated oxidative stress related genes. It indicated that BHPF exposure interfered the preference of male and female in courtship, and induced detrimental effects on reproduction. BHPF treatment decreased locomotor activity and time spent in top, increased freezing bouts, and induced anxiety/depression-like behavior. The tyrosine hydroxylase in brain decreased under BHPF exposure. Here we showed the potential adverse effects of BHPF on reproduction and exploratory behaviors.

Rongchang and Xifeng capsules suppress pentylenetetrazole-induced seizures and change rest/wake behavior in zebrafish larvae.

OBJECTIVE: To evaluate the effects of Rongchang capsule and Xifeng capsule on pentylenetetrazole- induced epilepsy in zebrafish larvae and to explore the possible mechanisms behind their actions. METHODS: We utilized a trajectory tracking system to monitor seizures in zebrafish larva to confirm that certain concentrations of Rongchang capsule and Xifeng capsule produce antiepileptic effects. c-fos expression was assessed by quantitative reverse transcription-polymerase chain reaction to validate the efficacy of the capsules. Rest/wake behavior and correlation analysis predicted the targets of Rongchang capsule and Xifeng capsule. RESULTS: Larval movement times and total distances traveled by zebrafish larvae experiencing pentylenetetrazole (PTZ)-induced seizures were decreased by valproate treatment. Rongchang (500 µg/mL) and Xifeng (200 µg/mL) rescued the epileptic behaviors and down-regulated c-fos expression in the brains of larvae, which indicated antiepileptic effects. The rest/wake behavioral profiles showed that Rongchang and Xifeng differentially decreased rest time at night and increased larval locomotor activities during the day. Based on correlation between the actions of the two capsules and known compounds, we predicted that they might change rest/wake behaviors by affecting serotonin, GABAergic and histamine signaling pathways. CONCLUSION: The efficacy of Rongchang capsule and Xifeng capsule in alleviating epilepsy-like behaviors and molecular responses was confirmed. Our study provides insight into the capsules' effect on epilepsy.

Visual Servoing of Wheeled Mobile Robots Without Desired Images.

This paper proposes a novel monocular visual servoing strategy, which can drive a wheeled mobile robot to the desired pose without a prerecorded desired image. Compared with existing methods that adopt the teaching pattern for visual regulation, this scheme can still work well in the situation that the desired image has not been previously acquired. Thus, with the aid of this method, it is more convenient for mobile robots to execute visual servoing tasks. Specifically, to deal with nonexistence of the desired image, the reference frame is craftily defined by taking advantage of visual targets and the planar motion constraint, and the pose estimation algorithm is designed for the mobile robot with respect to the reference frame. Then, an adaptive visual regulation controller is developed to drive the mobile robot to the intermediate frame, where the parameter updating law is constructed for the unknown feature height based on the concurrent learning framework. Stability analysis shows that regulation errors and height identification error can converge simultaneously. Afterwards, the mobile robot is driven to the metric desired pose with the identified feature height. Both simulation and experimental results are provided to validate the performance of this strategy.

Behavioural effect of low-dose BPA on male zebrafish: Tuning of male mating competition and female mating preference during courtship process.

The ubiquity of environmental pollution by endocrine disrupting chemicals (EDCs) such as bisphenol A (BPA) is progressively considered as a major threat to aquatic ecosystems worldwide. Numerous toxicological studies have proved that BPA are hazardous to aquatic environment, along with alterations in the development and physiology of aquatic vertebrates. However, generally, there is a paucity in knowledge of behavioural and physiological effects of BPA with low concentration, for example, 0.22 nM (50 ng/L) and 2.2 nM (500 ng/L). Here we show that treatment of adult male zebrafish (Danio rerio) with 7 weeks low-dose (0.22 nM-2.2 nM) BPA, resulted in alteration in histological structure of testis tissue and abnormality in expression levels of genes involved in testicular steroidogenesis. Furthermore, low-dose BPA treatment decreased the male locomotion during courtship; and was associated with less courtship behaviours to female but more aggressive behaviours to mating competitor. Interestingly, during the courtship test, we observed that female preferred control male to male under low-dose BPA exposure. Subsequently, we found that the ability of female to chose optimal mating male through socially mutual interaction and dynamics of male zebrafish, which was based on visual discrimination. In sum, our results shed light on the potential behavioural and physiological effect of low-dose BPA exposure on courtship behaviours of zebrafish, which could exert profound consequences on natural zebrafish populations.

Stacked Multilayer Self-Organizing Map for Background Modeling.

In this paper, a new background modeling method called stacked multilayer self-organizing map background model (SMSOM-BM) is proposed, which presents several merits such as strong representative ability for complex scenarios, easy to use, and so on. In order to enhance the representative ability of the background model and make the parameters learned automatically, the recently developed idea of representative learning (or deep learning) is elegantly employed to extend the existing single-layer self-organizing map background model to a multilayer one (namely, the proposed SMSOM-BM). As a consequence, the SMSOM-BM gains several merits including strong representative ability to learn background model of challenging scenarios, and automatic determination for most network parameters. More specifically, every pixel is modeled by a SMSOM, and spatial consistency is considered at each layer. By introducing a novel over-layer filtering process, we can train the background model layer by layer in an efficient manner. Furthermore, for real-time performance consideration, we have implemented the proposed method using NVIDIA CUDA platform. Comparative experimental results show superior performance of the proposed approach.