Shenyang Institute of Automation,Chinese Academy Of Sciences
Not a member yet
23582 research outputs found
Sort by
Actuator fault modeling and fault-tolerant tracking control of multi-vectored propeller aerostat
No full textFor a multi-vectored propeller aerostat with actuator faults, this study presents a fault-tolerant tracking control strategy, which includes fault modeling, observer, force estimation and tracking controller. Fault modeling considers the four types of faults of vectored propellers, namely, thrust offset, thrust efficiency loss, vectored angle offset and vectored angle stuck. Actuator faults can be determined from the fault observer, which identifies the thrust offset from the acceleration difference of the faulty aerostat with the ideal model. For tracking positions, a traditional PID controller is constructed with virtual control, compensated with the estimated fault force. The control allocation scheme is proposed to redistribute the available actuators in case faults occur. Simulation results of position tracking prove the effectiveness of the proposed strategy
Optogenetically engineered cell-based graphene transistor for pharmacodynamic evaluation of anticancer drugs
No full textRapid and effective cell-based pharmacodynamic evaluation in vitro is crucial for providing an experimental basis for the effectiveness and safety of drugs, and the existing cell-based methods for pharmacodynamics evaluation are usually invasive, dependent on chemical reagents, and/or unable to monitor the process in real time. Here, an optogenetically engineered cell-based graphene transistor is presented as a biosensor for the pharmacodynamic evaluation of anticancer drugs. The biosensor consists of a bare graphene transistor and optogenetically engineered cells as the gate terminal. The photoresponse of engineered cells regulates the output of the transistor and the increment pattern in the transistor output current upon drug action can be used to evaluate the drug efficacy. The results show that the optogenetic engineering of cancer cells does not affect the killing effect of drugs on the cells, and validate the effectiveness of the biosensor. The patterns of photoinduced increments exhibit significant variation with drug action time within 4 h or drug concentration in a range of 1 nM to 1 mM, and can qualitatively characterize the drug efficacy. Furthermore, the drug efficacy can be quantitatively evaluated with an indicator by logarithmically fitting the photoinduced increment patterns. The result also shows that the drain -source voltage significantly affects the evaluation performance and it is necessary to calibrate the voltage value to enhance the performance of the biosensor. The proposed biosensor affords simple, non-destructive, and time-efficient pharmacodynamic evaluation in vitro and is significant to understand the effect and mechanism of drugs in its early development stage
Towards a general design evaluation tool: The development and validation of a VPP for autonomous sailing monohulls
No full textSailing speed performance is a crucial indicator that significantly affects the trafficability, efficiency, and tracking capability of autonomous sailing monohulls during marine science missions. Considering that the design of the hull and keel of an autonomous sailing monohull is usually a task-orientated and creative process, estimating speed performance by traditional velocity prediction programs (VPPs) based on empirical formulas and gradient solvers will lead to errors. This paper proposes a generalized VPP for helping designers assess the speed performance of their autonomous sailing monohulls. We designed an enhanced genetic algorithm (GA) solver to help the VPP converge quickly without a priori performance estimation. Furthermore, we propose an innovative neighbourhood information-based optimization (NIBO) strategy to accelerate and refine the solutions using adjacent states (external conditions with the same true wind speed (TWS) or true wind angle (TWA)) instead of culminating prediction by solving each state independently. We provide an application of the proposed VPP on our prototype as an example. Moreover, the numerical and experimental results show that the proposed VPP can serve as a practical design evaluation tool, especially in the early stages of design
一种水下机器人用自适应坐底机构
No full text本发明涉及水下机器人技术领域，特别涉及一种水下机器人用自适应坐底机构。包括动力单元、坐底框架、主动摆杆、从动连杆及基座，其中基座设置于坐底框架的上方，基座通过多组从动连杆与坐底框架铰接；动力单元设置于基座上，主动摆杆的一端与坐底框架铰接，另一端与动力单元的输出端铰接，动力单元用于驱动主动摆杆摆动，从而带动基座进行升降动作。本发明具有自适应调节坐底高度功能，可针对不同的海底底质进行自适应调节，以防止被海流冲走或被底质吸附
一种驻车执行器生产线的电机组件装配设备
No full text本发明涉及驻车执行器生产领域，具体地说是一种驻车执行器生产线的电机组件装配设备，包括装配机构、旋转定位机构和电机组件搬运机构，其中装配机构包括PCB板定位组件、柱齿轮移动组件、第一压装组件和第二压装组件，PCB板定位组件设有可移动的定位杆，且PCB板通过所述定位杆插入对应的触点定位孔中实现定位，柱齿轮移动组件设有可移动的气动夹手，且柱齿轮通过所述气动夹手驱动移动至装配位置，旋转定位机构设有可升降的旋转头，且柱齿轮装配定位时，旋转头上升与柱齿轮接触并旋转直至插入柱齿轮中心孔中，电机组件通过电机组件搬运机构驱动移动至装配位置，并通过第一压装组件和第二压装组件完成压装。本发明装配效率大大提高，且保证装配质量
一种并联三体水下机器人稳心高度的调节方法
No full text本发明涉及一种并联三体水下机器人稳心高度的调节方法。本发明所提调节方法主要包括两个步骤。第一，调节并联三体水下机器人上下单体的浮力，通过改变上下单体的浮力差来调节其稳心高度。第二，调节并联三体水下机器人所构成等腰三角形的顶角大小，通过改变上下单体相对位置来进一步调节其稳心高度。本发明给出了浮力调节量、调节后顶角大小与稳心高度的具体函数关系，从而实现通过调节各单体浮力调节量、等腰三角构型的顶角大小，调节整体构型的稳心高度。本发明具有机理明确清晰、方法验证有效等特点，可广泛应用于并联多体水下机器人稳心高度的调节工作，以更好应对不同海流环境的干扰
自动化无人机连续发射系统
No full text本实用新型涉及一种自动化无人机连续发射系统，包括弹射器、调整平台、升降调节装置和控制系统，调整平台通过三个升降调节装置支撑，且所述升降调节装置下端安装于船体上，弹射器设于调整平台上，无人机设于弹射器上，各个升降调节装置的伸缩量通过所述控制系统控制，且所述控制系统根据船体运动计算升降调节装置的伸缩量。本实用新型利用调整平台补偿海浪引起的船体运动，从而保证无人机能够不受海浪影响顺利起飞，尤其适用于易受海浪影响的小型舰船，并且本实用新型的控制系统根据LQR最优控制理论对各个升降调节装置的升降伸缩量进行调节，不仅能够准确及时地调节所述调整平台位姿，同时整个过程自动完成，无需其他干预
基于多传感器数据融合浇注速度精确建模及在线检测方法
No full text本发明提供一种基于多传感器数据融合的浇注速度精确建模及在线检测方法，通过浇注速度与多传感器测得的物料粘度、挤压速度、阀门开度、挤压压力等在线参数建立对出口流速的预测模型，并通过实验数据的不断修正及验证实现出口流速的在线精准检测。针对在FDJ壳体内无法布置传感器进行直接测量的浇注速度及浇注量关键工艺参数，基于数据分析建模的方法，通过机器学习技术，利用可直接测量的工艺参数、设备运行参数等大量试验数据构建浇注速度的在线预测模型，实现其间接在线检测。通过实验数据的拟合、校正及验证实现出口流速及浇注量的在线精准检测，可以有效提高加工效率，并且满足工艺精度控制的要求
A fast tube model predictive control scheme based on sliding mode control for underwater vehicle-manipulator system
No full textThis paper presents a novel robust fast tube model predictive control (FTMPC) scheme for controlling the multi-input multi-output (MIMO) underwater vehicle-manipulator system (UVMS) in the case of system unmodeled uncertainties and external disturbances. The structure of the FTMPC is designed by a nominal fast MPC and an ancillary tube sliding mode control (SMC) law. The nominal fast MPC uses the approximated prediction control law for achieving the fast online convex optimization. The ancillary tube law uses the explicit SMC for strong robustness and makes the state error between the actual system and the nominal system converge to zero. Finally, the effectiveness of controlling UVMS is verified by the proposed SMC-based FTMPC scheme through the tracking simulations.</p
A transfer weighted extreme learning machine for imbalanced classification
No full textPrevious class imbalance learning methods are mostly grounded on the assumption that all training data have been labeled, however, is impractical in many real-world applications. The limited amount of labeled instances may produce a classifier with poor generalization. To address the issue, a transfer weighted extreme learning machine (TWELM) classifier is proposed, with the purpose of extracting knowledge from other domains to improve the classification performance of a classifier in a limited labeled target domain. To be specific, a well-tuned weighted extreme learning machine classifier is first learned from source data that has been completely labeled. Subsequently, another extreme learning machine classifier is obtained from the limited labeled target domain data to preserve the target domain structural knowledge and the decision boundary information. Finally, the target classifier is optimized by minimizing the outputs of the two classifiers on unlabeled target data. Experimental results on real-world data sets show that TWELM outperforms existing algorithms on classification accuracy and computation cost