Showing papers in "IEEE-ASME Transactions on Mechatronics in 2021"
TL;DR: In this article, a novel feature extraction method is proposed to extract health indicators (HIs) from general discharging conditions, and typical data-driven methods, including linear regression, support vector machine, relevance vector machine and Gaussian process regression (GPR), are constructed to predict battery SOH.
Abstract: State of health (SOH) is essential for battery management, timely maintenance, and safety incident avoidance. For specific applications, a variety of SOH estimation methods have been proposed. However, it is often difficult to apply these methods to other applications. In this article, a novel feature extraction method is proposed to extract health indicators (HIs) from general discharging conditions. A voltage partition strategy is used to obtain the discharge capacity differences of two cycles [△ Q ( V )] from nonmonotonic or pulse discharge voltage curve, and a filtering strategy is employed to obtain smooth voltage curves under dynamic discharging conditions. The standard deviations of the discharge capacity curve and △ Q ( V ) are selected as HIs and are verified to have strong correlations to battery capacity under different datasets for three types of batteries. By using these HIs as input features, typical data-driven methods, including linear regression, support vector machine, relevance vector machine, and Gaussian process regression (GPR), are constructed to predict battery SOH. The estimation results of these methods are compared under different operating conditions for the three types of batteries. Good estimation accuracy is achieved for all these methods. Among them, the GPR has the best performance, and its maximum absolute error and root-mean-square error are lower than 1% and 1.3%, respectively.
128 citations
TL;DR: A novel deep transfer learning model is constructed based on an adversarial learning strategy, which can effectively separate multiple unlabeled new fault types from labeled known ones and suggest that it is promising to address fault diagnosis transfer tasks in which the multiple new faults occur in the target domain.
Abstract: Recently, deep transfer learning based intelligent fault diagnosis has been widely investigated, and the tasks that source and target domains share the same fault categories have been well addressed. However, due to complexity and uncertainty of mechanical equipment, unknown new faults may occur unexpectedly. This problem has received less attention in the current research, which seriously limited the application of deep transfer learning. In this article, a two-stage transfer adversarial network is proposed for multiple new faults detection of rotating machinery. First, a novel deep transfer learning model is constructed based on an adversarial learning strategy, which can effectively separate multiple unlabeled new fault types from labeled known ones. Second, an unsupervised convolutional autoencoders model with silhouette coefficient is built to recognize the number of new fault types. Extensive experiments on a gearbox dataset validate the practicability of the proposed scheme. The results suggest that it is promising to address fault diagnosis transfer tasks in which the multiple new faults occur in the target domain, which greatly expand the application of deep transfer learning.
106 citations
TL;DR: A modified stacked auto-encoder that uses adaptive Morlet wavelet is proposed to automatically diagnose various fault types and severities of rotating machinery and experimental results show that the proposed method is superior to other state-of-the-art methods.
Abstract: Intelligent fault diagnosis techniques play an important role in improving the abilities of automated monitoring, inference, and decision-making for the repair and maintenance of machinery and processes In this paper, a modified stacked auto-encoder (MSAE) that uses adaptive Morlet wavelet is proposed to automatically diagnose various fault types and severities of rotating machinery Firstly, the Morlet wavelet activation function is utilized to construct an MSAE to establish an accurate nonlinear mapping between the raw nonstationary vibration data and different fault states Then, the nonnegative constraint is applied to enhance the cost function to improve sparsity performance and reconstruction quality Finally, the fruit fly optimization algorithm (FOA) is used to determine the adjustable parameters of the Morlet wavelet to flexibly match the characteristics of the analyzed data The proposed method is used to analyze the raw vibration data collected from a sun gear unit and a roller bearing unit Experimental results show that the proposed method is superior to other state-of-the-art methods
86 citations
TL;DR: In this article, a random forest-based classification framework was proposed to quantify the importance and correlations of battery manufacturing features and their effects on the classification of electrode properties. But, the proposed framework is not suitable for the analysis of battery electrodes.
Abstract: Lithium-ion battery manufacturing is a highly complicated process with strongly coupled feature interdependencies, a feasible solution that can analyse feature variables within manufacturing chain and achieve reliable classification is thus urgently needed. This article proposes a random forest (RF)-based classification framework, through using the out of bag (OOB) predictions, Gini changes as well as predictive measure of association (PMOA), for effectively quantifying the importance and correlations of battery manufacturing features and their effects on the classification of electrode properties. Battery manufacturing data containing three intermediate product features from the mixing stage and one product parameter from the coating stage are analysed by the designed RF framework to investigate their effects on both the battery electrode active material mass load and porosity. Illustrative results demonstrate that the proposed RF framework not only achieves the reliable classification of electrode properties but also leads to the effective quantification of both manufacturing feature importance and correlations. This is the first time to design a systematic RF framework for simultaneously quantifying battery production feature importance and correlations by three various quantitative indicators including the unbiased feature importance (FI), gain improvement FI and PMOA, paving a promising solution to reduce model dimension and conduct efficient sensitivity analysis of battery manufacturing.
79 citations
TL;DR: A family of coordinate transformations are devised to convert the MASH system tracking error dynamics into translation–rotation cascade manners, whereby the heterogeneity is removed and finite-time observers for complex unknowns are facilitated.
Abstract: In this paper, for a marine aerial-surface heterogeneous (MASH) system composed by a quadrotor unmanned aerial vehicle (UAV) and an unmanned surface vehicle (USV) with heterogeneity, completely unknown dynamics and disturbances, the accurate trajectory-tracking problem is solved by creating a novel coordinated trajectory tracking control (CTTC) scheme. A family of coordinate transformations are devised to convert the MASH system tracking error dynamics into translation-rotation cascade manners whereby the heterogeneity is removed and finite-time observers for complex unknowns are facilitated. In conjunction with sliding mode-based rotation error dynamics, distributed tracking controllers for the quadrotor UAV and the USV are independently synthesized such that cascade tracking error dynamics are globally asymptotically stable. With the aid of cascade and Lyapunov analysis, the entire CTTC solution to the accurate trajectory-tracking problem of the MASH system is eventually put forward. Simulation results and comprehensive comparisons on a prototype MASH system demonstrate the effectiveness and superiority of the proposed CTTC scheme.
64 citations
TL;DR: A novel hierarchical framework for the flexible motion of the six wheel-legged robot (BIT-6NAZA) is considered, indicating that it is a superior case of a selectable flexible motion with satisfactory stable performance under the field world environment.
Abstract: In complex real-world scenarios, wheel-legged robots with maneuverability, stability and reliability have addressed growing research attention, especially in material transportation, emergency rescue, as well as the exploration of unknown environments. How to achieve stable high-level movement with payload delivery simultaneously is the main challenge for the wheel-legged robot. In this paper, a novel hierarchical framework for the flexible motion of the six wheel-legged robot is considered in experimental results. Firstly, for the wheeled motion, the speed consensus algorithm is implemented to the six-wheeled cooperative control; for the legged motion, three gait sequences and foot-end trajectory based on the Bezier function are designed. Furthermore, a whole-body control architecture includes the attitude controller, impedance controller and center height controller is developed for obstacle avoidance, which can ensure the horizontal stability of the body of the robot when it passes through obstacles in different terrain. Finally, extensive experimental demonstrations using the six wheel-legged robot (BIT-6NAZA) are dedicated to the effectiveness and robustness of the developed framework, indicating that it is a superior case of a selectable flexible motion with satisfactory stable performance under the field world environment.
61 citations
TL;DR: An improved deadbeat predictive stator flux control (DPSFC) based on disturbance observer is proposed to improve the control performance of in-wheel permanent magnet synchronous motors (PMSMs) with parameter mismatch and disturbance.
Abstract: In this paper, an improved deadbeat predictive stator flux control (DPSFC) based on disturbance observer is proposed to improve the control performance of in-wheel permanent magnet synchronous motors (PMSMs) with parameter mismatch and disturbance. First, the sensitivity of conventional deadbeat predictive current control to the parameter variation, including flux linkage, stator resistance and stator inductance, is analyzed. Then, a reduced-order observer based on additional disturbance state variables is designed to predict the future stator flux and observe the system disturbance caused by parameter mismatch. The proposed DPSFC method is able to enhance the robustness of the drive performance effectively via the compensations of one-step delay and stator voltage. Finally, the performance of the proposed control method is validated by simulations and experiments on a prototype of an in-wheel PMSM drive.
59 citations
TL;DR: Three acceleration-level joint-drift-free ALJDF schemes for kinematic control of redundant manipulators are proposed and analyzed from perspectives of dynamics and kinematics with the corresponding tracking error analyses to enhance the product quality and production efficiency in industrial production.
Abstract: In this article, three acceleration-level joint-drift-free (ALJDF) schemes for kinematic control of redundant manipulators are proposed and analyzed from perspectives of dynamics and kinematics with the corresponding tracking error analyses. First, the existing ALJDF schemes for kinematic control of redundant manipulators are systematized into a generalized acceleration-level joint-drift-free scheme with a paradox pointing out the theoretical existence of the velocity error related to joint drift. Second, to remedy the deficiency of the existing solutions, a novel acceleration-level joint-drift-free (NALJDF) scheme is proposed to decouple Cartesian space error from joint space with the tracking error theoretically eliminated. Third, in consideration of the uncertainty at the dynamics level, a multi-index optimization acceleration-level joint-drift-free scheme is presented to reveal the influence of dynamics factors on the redundant manipulator control. Afterwards, theoretical analyses are provided to prove the stability and feasibility of the corresponding dynamic neural network with the tracking error deduced. Then, computer simulations, performance comparisons, and physical experiments on different redundant manipulators synthesized by the proposed schemes are conducted to demonstrate the high performance and superiority of the NALJDF scheme and the influence of dynamics parameters on robot control. This work is of great significance to enhance the product quality and production efficiency in industrial production.
51 citations
TL;DR: A practical terminal sliding mode control framework based on an adaptive disturbance observer (ADO) is presented for the active suspension systems and a detailed comparison with the active disturbance rejection method has been provided.
Abstract: In this article, a practical terminal sliding mode control (TSMC) framework based on an adaptive disturbance observer (ADO) is presented for the active suspension systems. The proposed controller requires no exact feedback linearization about the suspension dynamics. The ADO is designed to estimate the unknown dynamics and control errors produced by the motor actuator. To guarantee the fast convergence and high control accuracy, a TSMC-type surface and a continuous sliding mode reaching law are designed. The finite-time convergence of the controlled system is guaranteed based on the Lyapunov stability theory. To evaluate the performance improvement of the proposed control framework, a detailed comparison with the active disturbance rejection method has been provided. Finally, a practical hardware-in-loop experiment is implemented to validate the effectiveness of the proposed control scheme.
50 citations
TL;DR: A novel recursive robust integral of the sign of the error (RISE) control method is proposed for mechanical servosystems with mismatched uncertainties and can theoretically achieve remarkable asymptotic tracking performance with zero steady-state error in spite of matched and mismatched time-variant uncertainties.
Abstract: Uncertainties, especially mismatched uncertainties, pose great challenges to high accuracy tracking controller design for mechanical servosystems. In this article, a novel recursive robust integral of the sign of the error (RISE) control method is proposed for mechanical servosystems with mismatched uncertainties. In the controller development, two auxiliary error signals are introduced into the recursive backstepping design framework, and then, RISE feedbacks are synthesized to eliminate the matched and mismatcheduncertainties simultaneously. Moreover, to reduce the design conservatism, an adaptive recursive RISE control law is also developed for mechanical servosystems suffering from both parametric uncertainties and unmodeled disturbances, in which desired-trajectory-based adaptation law is synthesized to achieve compensation for parametric uncertainties. The proposed control methods can theoretically achieve remarkable asymptotic tracking performance with zero steady-state error in spite of matched and mismatched time-variant uncertainties. The proposed controllers are applied to an actual hydraulic servosystem and comparative experiments are performed to verify their effectiveness.
49 citations
TL;DR: In this paper, a cross-coupled second-order discrete-time fractional-order sliding mode control strategy is proposed to reduce the incoordination among driving linear motors.
Abstract: This article presents a universal method of precise synchronization control for linear-motor-driven systems. The control method named cross-coupled second-order discrete-time fractional-order sliding mode control contains a cross-coupled control strategy to reduce the incoordination among driving linear motors. It also includes the second-order structure and fractional-order sliding mode surface to reduce the chattering phenomenon and improve the dynamic performance simultaneously, so that the precision is further enhanced. In particular, a universal definition of synchronization error is presented so that a better synchronization control performance can be achieved, especially in multidimensional systems. It is also compatible well with the contouring control, which extends the application further. Moreover, the stability of the controller is analyzed in this article. Finally, the proposed control method is conducted in simulations and experiments under various tasks, whose results have proved its effectiveness and advantages over conventional methods.
TL;DR: This article presents a new variable curvature kinematic modeling approach for soft continuum robots by taking the external forces into consideration, achieving both accurate motion simulation and feedforward control of the robot.
Abstract: The compliant structure and influence of external forces usually result in complex deformation of soft continuum robots, which makes the accurate modeling and control of the robot challenging In this work, we present a new variable curvature kinematic modeling approach for soft continuum robots by taking the external forces into consideration, achieving both accurate motion simulation and feedforward control of the robot To this end, the variable curvature configuration is firstly parameterized based on the absolute nodal coordinate formulation (ANCF) Then, a kinematic model is developed to describe the mappings between the defined configuration space and the actuation space with payloads With this model, we achieve accurate and fast motion simulation for the soft continuum robot with different payloads and input pressures within 1 millisecond, which is verified by a set of experiments Finally, an inverse-model-based feedforward controller is developed for a two-section soft continuum robot The experimental results of tracking complex trajectories verify the effectiveness of our model and control strategies The average position error of the end-effector is 289% of the robot length This work can also be served as a tool to design and analyze soft continuum robots with desired workspace
TL;DR: In this article, a quadruped quadruped piezoelectric actuator is used to achieve nanopositioning in a large travel range, and the operating principle of combining bionic walking and swinging actuation modes is illustrated and a dynamic model and control method is developed.
Abstract: A piezoelectric platform driven by a bionic quadruped piezoelectric actuator is developed. It uses the operating principle of combining bionic walking and swinging actuation modes to achieve nanopositioning in a large travel range. The mechanical structure is introduced, the operation principle is illustrated, and the dynamic model and control method are developed. The open-loop performances are first tested to verify the effectiveness of the operating principle and the dynamic model, and the simulation results agree well with the experimental results. Then, the closed-loop experiments in bionic walking and swinging actuation modes are carried out, respectively, and the controllers are designed based on the dynamic model. In the point-to-point positioning control experiments, the steady-state errors for the target position of ±1000 μ m in axes X and Y are within ±1 μ m in bionic walking actuation mode, and they are within ±20 nm for the target position of ±2 μ m in bionic swinging actuation mode. The closed-loop experiments under the combination of the bionic walking actuation mode and swinging actuation mode are performed, and a switched controller is developed to obtain the switching of the two modes automatically; the steady-state errors are within ±20 nm. The experimental results confirm that the method combining bionic walking and swinging actuation modes and switched control is valid in enhancing the positioning precision and travel range for the nanopositioning platform.
TL;DR: A federated transfer learning method for fault diagnosis with federal initialization stage to keep similar data structures in distributed feature extractions, and a federated communication stage is further implemented using deep adversarial learning.
Abstract: Intelligent data-driven machinery fault diagnosis methods have been popularly developed in the past years. While fairly high diagnosis accuracies have been obtained, large amounts of labeled training data are mostly required, which are difficult to collect in practice. The promising collaborative model training solution with multiple users poses high demands on data privacy due to conflict of interests. Furthermore, in the real industries, the data from different users can be usually collected from different machine operating conditions. The domain shift phenomenon and data privacy concern make the joint model training scheme quite challenging. To address this issue, a federated transfer learning method for fault diagnosis is proposed in this study. Different models can be used by different users to enhance data privacy. A federal initialization stage is introduced to keep similar data structures in distributed feature extractions, and a federated communication stage is further implemented using deep adversarial learning. A prediction consistency scheme is also adopted to increase model robustness. Experiments on two real-world datasets suggest the proposed federated transfer learning method is promising for real industrial applications.
TL;DR: A novel human-in-the-loop control framework for a fully actuated lower limb exoskeleton with high degree-of-freedoms (DoFs), allowing users to walk without crutches or other external stabilization tools is proposed.
Abstract: Exoskeletons are increasingly used to assist humans in military, industry, and healthcare applications, thereby enabling individuals to gain increased strength and endurance. This article proposes a novel human-in-the-loop control framework for a fully actuated lower limb exoskeleton with high degree-of-freedoms (DoFs), allowing users to walk without crutches or other external stabilization tools. To imitate the natural lower limb motion of users, a novel barrier energy function is utilized for the design of the control strategy, where the human-robot manipulation space is reformulated as a human-voluntary and a robot-constrained region. The variations in the barrier energy function are based on the distance between the center of mass and zero moment point of the walking exoskeleton, thereby constraining the lower limb motion of the user to a compliant region around various desired trajectories. Based on varying regional functions, the proposed strategy is designed to control the exoskeleton to follow appropriate ergonomic trajectories. For such a purpose, an adaptive controller is exploited considering the functions of the human effort and the robot's capabilities simultaneously, and a smooth motion transition can be achieved between the human and robot regions. Finally, physical experiments are conducted on a ten-DoFs walking exoskeleton to validate the stability and robustness of the proposed control framework with subjects performing flat walking, turning, and obstacle avoidance movements.
TL;DR: A novel vision-based cutting control algorithm is proposed to cut a deformable object along the predesigned path with specified cutting depth to model soft tissue considering viscoelasticity and asymptotic stability is proved by Lyapunov analysis.
Abstract: Automatic cutting is an essential task in the field ofrobot-assisted surgery. In this article, a novel vision-based cutting control algorithm is proposed to cut a deformable object along the predesigned path with specified cutting depth. The method to model soft tissue considering viscoelasticity is developed, and unknown parameters of the deformation model are estimated online by introducing the visual feedback of feature points. According to the position of trajectory points after deformation, we illustrate how to generate the desired pose of the knife that avoids large deformation. In this way, the cutting task has been recast into a visual tracking problem, that is, controlling the knife's projection to track a target surface in the image plane. To cope with this problem, we choose two parallel linear segments (including the direction and the length) extracted from the edge of the knife's projection as image features, which has a one-to-one mapping with the pose of the knife, and design a dynamic-based controller based on the combined feature. The asymptotic stability of the closed-loop system is proved by Lyapunov analysis. A series of experiments using different materials are conducted to validate the effectiveness of the proposed cutting control algorithm.
TL;DR: In this paper, a robust nonlinear model predictive control (NMPC) scheme is proposed for the visual servoing of quadrotors subject to external disturbances, where the image moments are defined in the virtual camera plane and adopted as visual features to derive the decoupled image kinematics.
Abstract: In this article, a robust nonlinear model predictive control (NMPC) scheme is proposed for the visual servoing of quadrotors subject to external disturbances. By using the virtual camera approach, the image moments are defined in the virtual camera plane and adopted as visual features to derive the decoupled image kinematics. As a result, the image-based visual servoing (IBVS) system model is established by integrating the image kinematics and quadrotor dynamics. To handle the visibility constraint, a robust NMPC scheme is developed for the IBVS of the quadrotor such that the visual target can stay within the field of view of the camera. In addition, based on the Lipschitz condition, the tightened state constraints are constructed to tackle external disturbances. The sufficient conditions on guaranteeing recursive feasibility of the proposed NMPC algorithm are derived. Furthermore, we theoretically show that the tracking error will converge to a small set around the origin in finite time under some derived conditions. Finally, simulation studies and experimental tests are conducted to verify the efficacy of the proposed method.
TL;DR: This article addresses neurodynamics-based model predictive control of continuous-time under-actuated mechatronic systems by using a collaborative neurodynamic approach and the closed-loop system is proven to be asymptotically stable.
Abstract: This article addresses neurodynamics-based model predictive control of continuous-time under-actuated mechatronic systems. The control problem is formulated as a global optimization problem based on sampled data, which is solved by using a collaborative neurodynamic approach. The closed-loop system is proven to be asymptotically stable. Specific applications on control of autonomous surface vehicles and unmanned wheeled vehicles are elaborated to substantiate the efficacy of the approach.
TL;DR: The design and validation of a backdrivable powered knee orthosis for partial assistance of lower-limb musculature is presented, which aims to facilitate daily activities in individuals with musculoskeletal disorders.
Abstract: This paper presents the design and validation of a backdrivable powered knee orthosis for partial assistance of lower-limb musculature, which aims to facilitate daily activities in individuals with musculoskeletal disorders. The actuator design is guided by design principles that prioritize backdrivability, output torque, and compactness. First, we show that increasing the motor diameter while reducing the gear ratio for a fixed output torque ultimately reduces the reflected inertia (and thus backdrive torque). We also identify a tradeoff with actuator torque density that can be addressed by improving the motor's thermal environment, motivating our design of a custom Brushless DC motor with encapsulated windings. Finally, by designing a 7:1 planetary gearset directly into the stator, the actuator has a high package factor that reduces size and weight. Benchtop tests verify that the custom actuator can produce at least 23.9 Nm peak torque and 12.78 Nm continuous torque, yet has less than 2.68 Nm backdrive torque during walking conditions. Able-bodied human subjects experiments (N=3) demonstrate reduced quadriceps activation with bilateral orthosis assistance during lifting-lowering, sit-to-stand, and stair climbing. The minimal transmission also produces negligible acoustic noise.
TL;DR: A novel strategy is designed for trajectory control of a multisection continuum robot in three-dimensional space to achieve accurate orientation, curvature, and section length tracking and could manage simultaneous extension/contraction, bending, and torsion actions on mult isection continuum robots with decent performance.
Abstract: Despite the rise of development in continuum manipulator technology and application, a model-based feedback closed-loop control appropriate for continuum robot designs has remained a significant challenge. Complicated by the soft and flexible nature of the manipulator body, control of continuum structures with infinite dimensions proves to be difficult due to their complex dynamics. In this article, a novel strategy is designed for trajectory control of a multisection continuum robot in three-dimensional space to achieve accurate orientation, curvature, and section length tracking. The formulation connects the continuum manipulator dynamic behavior to a virtual discrete-jointed robot whose degrees of freedom are directly mapped to those of a continuum robot section under the hypothesis of constant curvature. Based on this connection, a computed torque control architecture is developed for the virtual robot, for which kinematics and dynamic equations are constructed and exploited, with appropriate transformations developed for implementation on the continuum robot. The control algorithm is implemented on a six degree-of-freedom two-section OctArm continuum manipulator. Experiments show that the proposed method could manage simultaneous extension/contraction, bending, and torsion actions on multisection continuum robots with decent performance (arc length and curvature error of ±4 mm and ±0.35 m–1). The designed dynamic controller can reduce the curvature tracking error and rise time by up to 48.1% and 94.8% compared to the traditional proportional-integral-derivative controller during two-section maneuvers.
TL;DR: In this paper, a generalized bumpless transfer concept is presented for switched LPV systems to describe the transient performance, and a family of time-driven switching controllers with bumpless constraint and fault-tolerant requirement are also designed.
Abstract: For switched LPV systems with possible actuator failures, the parameter-dependent multiple piecewise Lyapunov function is constructed to handle the $H_\infty$ bumpless transfer fault-tolerant control problem. First, a generalized bumpless transfer concept is presented for switched LPV systems to describe the transient performance, for which only the local bumpless transfer condition is required. Second, an event-triggered switching law, depending on the system states, external parameters and dwell time, is designed to ensure a time span among adjacent switchings. Third, a solvability condition of the $H_\infty$ bumpless transfer fault-tolerant control problem is developed. A family of time-driven switching controllers with bumpless transfer constraint and fault-tolerant requirement are also designed. Finally, an application example of an aero-engine is given to verify the effectiveness of the developed methods.
TL;DR: A novel hierarchical collaborative probabilistic semantic mapping framework is proposed, where the problem is formulated in a distributed setting and the modeling of the hierarchical semantic map fusion framework and its mathematical derivation of its probability decomposition is modeled.
Abstract: Performing collaborative semantic mapping is a critical challenge for cooperative robots to enhance their comprehensive contextual understanding of the surroundings. This article bridges the gap between the advances in collaborative geometry mapping that relies on pure geometry information fusion, and single robot semantic mapping that focuses on integrating continuous raw sensor data. In this article, a novel hierarchical collaborative probabilistic semantic mapping framework is proposed, where the problem is formulated in a distributed setting. The key novelty of this work is the modeling of the hierarchical semantic map fusion framework and its mathematical derivation of its probability decomposition. At the single robot level, the semantic point cloud is obtained by combining information from heterogeneous sensors and used to generate local semantic maps. At the collaborative robots level, local maps are shared among robots for global semantic map fusion. Since the voxel correspondence is unknown between local maps, an expectation-maximization approach is proposed to estimate the hidden data association. Then, Bayesian rule is applied to perform semantic and occupancy probability update. Experimental results on the unmanned aerial vehicle and the unmanned ground vehicle platforms show the high quality of global semantic maps, demonstrating the accuracy and utility in practical missions.
TL;DR: This work proposes a robotic lidar sensor based on incommensurable scanning that allows straightforward mass production and adoption in autonomous robots and features a peaked central angular density, enabling in applications that prefers eye-like attention.
Abstract: High performance lidars are essential in autonomous robots such as self-driving cars, automated ground vehicles and intelligent machines. Traditional mechanical scanning lidars offer superior performance in autonomous vehicles, but the potential mass application is limited by the inherent manufacturing difficulty. We propose a robotic lidar sensor based on incommensurable scanning that allows straightforward mass production and adoption in autonomous robots. Some unique features are additionally permitted by this incommensurable scanning. Similar to the fovea in human retina, this lidar features a peaked central angular density, enabling in applications that prefers eye-like attention. The incommensurable scanning method of this lidar could also provide a much higher resolution than conventional lidars which is beneficial in robotic applications such as sensor calibration. Examples making use of these advantageous features are demonstrated.
TL;DR: The results show that the proposed approach is able to make safe and personalized decisions, and execute motion control more efficiently for automated driving under dynamic situations, validating its feasibility and effectiveness.
Abstract: In this article, a novel approach of decision-making and motion control is designed for realizing safe and personalized driving of autonomous vehicles. A new lane-change intention generation model and a new lane-change decision-making algorithm are proposed. The feature of the proposed decision-making module is that the interactions between the ego vehicle and other surrounding vehicles are represented by the dynamic potential field (DPF) and embedded in the gap acceptance model to ensure the safety and personalization during driving. In addition, an integrated trajectory planning and tracking control algorithm, which incorporates the artificial potential field and constrained Delaunay triangulation (CDT) into the model predictive control framework, is developed. The newly developed integrated controller allows efficient execution of the expected motion. The proposed approach is tested under different driving conditions and further compared with an existing baseline method. The results show that the proposed approach is able to make safe and personalized decisions, and execute motion control more efficiently for automated driving under dynamic situations, validating its feasibility and effectiveness.
TL;DR: The development of a BSP-based CNN model for fault diagnosis and the extensive evaluation of CNN-TL methods for monitoring and diagnosing planetary gearboxes are described.
Abstract: To improve the efficiency and accuracy of fault diagnostics of planetary gearboxes, an intelligent diagnosis approach is proposed based on deep convolutional neural networks (CNNs) and vibration bispectrum (BSP). Rather than using raw vibration signals, BSP is appreciated as the input for the CNN models (denoted as BSP-CNN) because the BSP allows nonlinear feature enhancement and noise reduction. In addition, transfer learning (TL) is accompanied to address the challenges of CNN difficulties. The proposed BSP-CNN is verified firstly to diagnose a number of common faults including gear states: normal, tooth wear, tooth root crack, tooth breakage and missing tooth, achieving an accuracy of 97.36% in identifying different faults. Then, its TL capability is evaluated based on the sun gear faults datasets. The classification accuracy of the planet gear faults is over 95.1%. After the transfer learning, the classification accuracy of the sun gear fault is still higher than 97.9%, and the computational time consumed by proposed method is also less compared to other diagnosis methods. This article has twofold contributions: first, the development of a BSP-based CNN model for fault diagnosis; andsecond, the extensive evaluation of CNN-TL methods for monitoring and diagnosing planetary gearboxes.
TL;DR: The result shows that with only two fingers, the gripper can reliably grasp objects with various shapes and volume, especially small and fragile objects.
Abstract: This article presents a two-finger soft-robotic gripper with enveloping grasping (EG) and pinching grasping (PG) modes. The proposed soft-robotic gripper is based on a two-finger design that combines two dual-module pneumatic actuators with a variable chamber height (DMVCHA). The prototype of the DMVCHA has been fabricated by molding of silicone rubber. In the PG mode, the high passive compliance of the DMVCHA allows large contact area and realizes a vertically plane contact with the object, which improves the grasping reliability and is suitable for grasping objects with a small and medium size. In the EG mode, the DMVCHA becomes the pneumatic network actuator (PneuNet) with a variable chamber height, which provides greater grasping force and is suitable for grasping objects with larger and hollow size. Compared with the traditional PneuNets, the DMVCHA has a greater output force. The bending angle and output force of actuators with different structures have been analyzed by finite-element analysis. The grasping performance of the proposed and universal soft grippers is studied by experiments. The result shows that with only two fingers, the gripper can reliably grasp objects with various shapes and volume, especially small and fragile objects.
TL;DR: An unsupervised domain adaptation approach is developed to mitigate the domain shifts between the data gathered from the experimental platform and the operating platform by aligning the features extracted from the two data domains.
Abstract: In this article, the problem of the cross-domain fault diagnosis of rotating machinery is considered. In a practical setting of this approach, the operating platform of the machine may have a different setup and conditions compared to the experimental platform that is used to collect the training data. This can lead to significant data variations, specifically domain shifts. Conventional data-driven approaches are known to adapt poorly to these domain shifts, resulting in a significant drop in the diagnosis accuracy when the pretrained model is applied in the actual operating situation. In this article, an unsupervised domain adaptation approach is developed to mitigate the domain shifts between the data gathered from the experimental platform (the source domain) and the operating platform (the target domain) by aligning the features extracted from the two data domains. The mutual information between the target feature space and the entire feature space is maximized to improve the knowledge transferability of the labeled data in the source domain. Furthermore, the feature-level discrepancy between the two domains is minimized to further improve diagnosis accuracy. The experiments using public datasets and real-world adaptation scenarios demonstrate the feasibility and the superior performance of the proposed method.
TL;DR: With the novel design and control, the proposed aerial vehicle is capable of achieving superior maneuverability, better stability and motion accuracy in the presence of uncertainties, as well as improved power efficiency compared to traditional UAVs when performing dexterous aerial locomotion and manipulation.
Abstract: In this article, a novel multirotor unmanned aerial vehicle (UAV) is proposed for enhanced locomotion and manipulation in unstructured environments. The vehicle has a tilting-rotor architecture. Specifically, two pairs of rotors are mounted on two independently controlled tilting arms placed at two sides of the vehicle, forming an “H” configuration. Such a structure endows the vehicle with more degrees of freedom, improving the maneuverability without sacrificing energy efficiency at the cost of only two additional servos compared to traditional quadcopters. Based on this architecture, a dual-level adaptive robust control is developed to cope with inertial parametric uncertainties and uncertain nonlinearities for accurate motion tracking. To resolve the redundancy in actuation, a thrust force optimization problem minimizing power consumption while achieving the desired body force wrench is formulated and solved precisely and efficiently. With the novel design and control, the proposed aerial vehicle is capable of achieving superior maneuverability, better stability and motion accuracy in the presence of uncertainties, as well as improved power efficiency compared to traditional UAVs when performing dexterous aerial locomotion and manipulation. To demonstrate the advantage of the proposed new UAV design and control in real applications, we conduct four challenging experiments on aerial locomotion and manipulation: circular trajectory tracking, passing through a narrow tunnel, picking up an object from a cluttered shelf, and aerial hole drilling. Experimental results validate the applicability of the proposed innovation.
TL;DR: Using the developed surgical system, accurate positioning and successful cutting of desired straight-line and curvilinear paths on saw-bone phantoms behind the cup with different densities are demonstrated.
Abstract: This article presents the development and experimental evaluation of a redundant robotic system for the less-invasive treatment of osteolysis (bone degradation) behind the acetabular implant during total hip replacement revision surgery. The system comprises a rigid-link positioning robot and a continuum dexterous manipulator (CDM) equipped with highly flexible debriding tools and a fiber Bragg grating (FBG) based sensor. The robot and the continuum manipulator are controlled concurrently via an optimization-based framework using the tip position estimation (TPE) from the FBG sensor as feedback. Performance of the system is evaluated on a setup that consists of an acetabular cup and saw-bone phantom simulating the bone behind the cup. Experiments consist of performing the surgical procedure on the simulated phantom setup. CDM TPE using FBGs, target location placement, cutting performance, and the concurrent control algorithm capability in achieving the desired tasks are evaluated. Mean and standard deviation of the CDM TPE from the FBG sensor and the robotic system are 0.50 and 0.18 mm, respectively. Using the developed surgical system, accurate positioning and successful cutting of desired straight-line and curvilinear paths on saw-bone phantoms behind the cup with different densities are demonstrated. Compared to the conventional rigid tools, the workspace reach behind the acetabular cup is 2.47 times greater when using the developed robotic system.