Hu Su

Chinese Academy of Sciences | CAS · Institute of Automation

Learning control has been an active topic of re- search for several decades, and is of theoretical, as well as practi- cal, significance. Current theories and developments in learning control are discussed. Following a brief introduction of the state as well as new progress on learning control, we give a detail re- view on the models and algorithms...

Acoustic holography (AH) is a technique with significant potential in realms, such as biomedicine, industry, and augmented reality. The implementation of acoustic holograms can be realized by a passive approach or active ones. Although the passive approach (by a 3D printer) can achieve high-quality acoustic field generation, it is constrained by hi...

As an essential component of noncontact manipulation, acoustic manipulation has achieved great success in multidisciplinary research and applications. Although acoustic tweezers have made advancements in manipulating particles in air, handling individual particles with high precision in water remains challenging and inadequately addressed due to th...

AudioVisual Source Localization (AVSL) aims to locate sounding objects within video frames given the paired audio clips. Existing methods predominantly rely on self-supervised contrastive learning of audiovisual correspondence. Without any bounding-box annotations, they struggle to achieve precise localization, especially for small objects, and suf...

Acoustic holography (AH) provides a promising technique for arbitrary acoustic field reconstruction, supporting many applications like robotic micro-nano manipulation, neuromodulation, volumetric imaging, and virtual reality. In AH, three-dimensional (3D) acoustic fields quantified with complex-valued acoustic pressures are reconstructed by virtue...

Deep-learning-based detection methods have been widely applied to industrial defect inspection. However, directly using vanilla detection methods fails to achieve satisfying performance due to the lack of identifiable features. In this paper, a novel attention-based multi-scale feature fusion method (AMFF) is proposed, aiming to enhance defect feat...

Noncontact trapping of micro objects has great application potential in fields like material science and biomedical engineering due to its label-freeness and biocompatibility. In this paper, an automated acoustic micro-particle trapping system implemented with phased transducer array (PTA) is prototyped. The system is incorporated with a stereo vis...

Phase-only acoustic holography is a fundamental and promising technique for contactless robotic manipulation. Through independently controlling phase-only hologram (POH) of phase array of transducers (PAT) and simultaneously driving each channel by sophisticated circuits, a certain acoustic field is dynamically generated in working medium (e.g., ai...

Moving object segmentation (MOS) is one of the essential functions of the vision system of all robots, including medical robots. Deep learning-based MOS methods, especially deep end-to-end MOS methods, are actively investigated in this field. Foreground segmentation networks (FgSegNets) are representative deep end-to-end MOS methods proposed recent...

The robot-task-sequencing planning problem is investigated in this paper, where multi-robot tasks with resource conflicts and precedence constraints are involved making the problem challenging and complex. An effective approach is proposed to minimize the production cycle time while handling the high complexity effectively. The approach decouples t...

Acoustic holography (AH) is a promising technique for precise noncontact micro-nano robotic manipulation. It encodes a three-dimensional (3D) acoustic field acting as a virtual end-effector into a two-dimensional (2D) hologram, whereby the desired acoustic field reconstruction is made possible. Most traditional methods to implement AH, such as 3D p...

Sound Source Localization (SSL) and approaching are essential capabilities for robots with auditory sensing. However, most existing methods for SSL only provide the Direction of Arrival (DoA) of the source, without the source distance, and cannot cope with time-varying number of sound sources. In this paper, a novel framework that integrates multi-...

The paper investigates the problem of visual tracking on moving object in nanomanipulation inside scanning electron microscope (SEM). Image noise is a primary concern when dealing with the problem, which includes the inherent statistical noise and that induced by the motion distortion. A visual tracking method with SEM image denoising algorithm is...

This article presents a study on trajectory tracking control for an underactuated bionic underwater vehicle that is propelled by undulating fins. The vehicle's propulsion mechanism is inspired by marine organisms, and its working principle is briefly introduced. Accurate trajectory tracking control is challenging due to the coupling of fin motions...

Disturbances caused by unknown dynamics and environmental factors render the automatic control of underwater vehicles extremely challenging. These effects are complex, time varying, and difficult to model accurately, leading to possible instability in the control process. This article focuses on the disturbance rejection problem of an underactuated...

Deep convolutional neural networks (DCNN) trained with pixel-level segmentation masks achieve high performance in the task of instance segmentation. The difficulty of acquiring such annotation limits the application and popularization of the DCNN-based approaches. To address the issue, a weakly supervised approach is proposed in the paper which per...

Anchor-free detectors basically formulate object detection as dense classification and regression. For popular anchor-free detectors, it is common to introduce an individual prediction branch to estimate the quality of localization. The following inconsistencies are observed when we delve into the practices of classification and quality estimation....

Anchor-free detectors basically formulate object detection as dense classification and regression. For popular anchor-free detectors, it is common to introduce an individual prediction branch to estimate the quality of localization. The following inconsistencies are observed when we delve into the practices of classification and quality estimation....

Deep Convolutional Neural Network (DCNN) has recently advanced state-of-the-art performance on vision-related tasks and its application is further extended to industrial fields. The paper focuses on the problem of fabric defect detection to which an efficient DCNN architecture is developed. In contrast to previous methods that directly apply existi...

Large-scale data with human annotations is of crucial importance for training deep convolutional neural network (DCNN) to ensure stable and reliable performance. However, accurate annotations, such as bounding box and pixel-level annotations, demand expensive labeling efforts, which has prevented wide application of DCNN in industries. Focusing on...

This paper investigates automatic quality inspection for the components with small diameter and deep aperture. An automatic pick-and-place system is constructed, which employs an endoscope to achieve better image quality aiming at the char-acteristics of the component. A coarse-to-fine contour extraction algorithm with four steps is presented to in...

In this paper, we propose an unsupervised background reconstruction method to detect defects on surfaces with unevenly distributed textures. An improved deep convolutional autoencoder is utilized to reconstruct the textured background of the original image as a defect-free reference. Specifically, a weighted loss function based on structural simila...

Both accuracy and efficiency are significant for pose estimation and tracking in videos. State-of-the-art performance is dominated by two-stages top-down methods. Despite the leading results, these methods are impractical for real-world applications due to their separated architectures and complicated calculation. This paper addresses the task of a...

Presents corrections to the footnote from the paper,“An efficient insertion control method for precision assembly of cylindrical components,” (Liu, S., et al), IEEE Trans. Ind. Electron., vol. 64, no. 12, pp. 9355–9365, Dec. 2017.

The paper focuses on the contour extraction for the inner wires of a kind of deep hole component to achieve a high-accuracy inspection. The vision system consisting of a cam-era and an endoscope is developed to acquire high-quality images of the internal structure. For the acquired images, a contour ex-traction method is proposed, which could be di...

Object contour plays an important role in fields such as semantic segmentation and image classification. However, the extraction of contour is a difficult task, especially when the contour is incomplete or unclosed. In this paper, the existing contour detection approaches are reviewed and roughly divided into three categories: pixel-based, edge-bas...

The emergency of surface defect would significantly influence the quality of MPCG (Mobile Phone Cover Glass). Therefore, efficient defect detection is highly required in the manufacturing process. Focusing on the problem, an automatic detection system is developed in this paper. The system adopts backlight imaging technology to improve the signal t...

The amplification products of 3’RACE (A) and 5’RACE (B). (TIF)

The full length of ICS cDNA in S. baicalensis roots. (TIF)

TSA content in S. baicalensis roots (S4-1) under stress and (S4-2) in exogenous SA treatment. (DOCX)

Baicalein content in S. baicalensis roots (S6-1) under stress and (S6-2) in exogenous SA treatment. (DOCX)

FSA content in S. baicalensis roots (S3-1) under stress and (S3-2) in exogenous SA treatment. (DOCX)

HPLC chromatogram of baicalin and baicalein in S. baicalensis roots in exogenous SA treatment. A-24 h, B-48 h, C-72 h; (i) Standard chromatogram, peak 1 is baicalin, peak 2 is baicalein. (ii) control, (iii) 70 μM SA, (iv) 140 μM SA, (v) 280 μM SA. (TIF)

Baicalin content in S. baicalensis roots (S5-1) under stress and (S5-2) in exogenous SA treatment. (DOCX)

HPLC chromatogram of baicalin and baicalein in S. baicalensis roots under drought and salt stress. (i) Standard chromatogram, peak 1 is baicalin, peak 2 is baicalein. (ii) control, (iii) drought stress, (iv) salt stress. (TIF)

Cinnamic acid concentration in PAL activity analysis (S1-1) under stress and (S1-2) in exogenous SA treatment. (DOCX)

Fluorescence value in ICS activity analysis (S2-1) under stress and (S2-2) in exogenous SA treatment. (DOCX)

In this paper, an efficient insertion control strategy is presented for precision assembly of cylindrical components with interference fit. It has two inherent advantages compared with existing insertion methods in the domain of precision assembly. First, it does not need preparatory actions dedicated for horizontal forces before insertion actions,...

In this paper, a path planning algorithm is proposed to search feasible paths for an auxiliary robot working in complex environment. The movement of the robot could thus be guided by the planned result of the algorithm in assembly practice. In the process of environmental modeling, a quad tree method is employed to construct the non-uniform environ...

For a large auxiliary robot working in complex environment, an improved ant colony algorithm based on non-uniform environment modeling is proposed to search feasible paths. In the process of environment modeling, convex projection is firstly conducted to calculate the obstacle area of each environment obstacle on the projection plane. Then, the obs...

In this paper, a model of computing the desired velocity of returning balls is developed so that the robot could return the incoming ball to a desired point on the table with specified landing velocity. At first, three polynomials of the flying time are applied to fit the ball's trajectory while neglecting the effort of Magnus force; one result of...

In this brief, we consider the problem of controlling the racket attached to the ping-pong playing robot, so that the incoming ball is returned to a desired position. The maps that are used to calculate the racket's initial parameters are described. They are implemented with the locally weighted regression (LWR). An active learning approach based o...

A model of computing the desired velocity of returning balls is developed so that the robot could return the incoming ball to a desired point on the table with a specified landing velocity. Firstly, three polynomials of the flying time are used to fit the ball's trajectory while neglecting the effort of the Magnus force; the initial velocity is obt...

An analytic flying model that can well represent the physical behavior is derived, where the ball's self-rotational velocity changes along with the flying velocity. Based on the least square method, a rebound model that represents the relation between the velocities before and after rebound is established. The initial trajectory is fitted to three...