The paper presents the proposition of a numerical model of the human circulatory system modification for the purpose of physical reproduction of conditions prevailing in the apex of the heart on a hybrid simulator. Th...
详细信息
The paper presents the proposition of a numerical model of the human circulatory system modification for the purpose of physical reproduction of conditions prevailing in the apex of the heart on a hybrid simulator. This is a physical-numerical stand which performs a generation of hydraulic conditions occurring in the selected points of circulatory system. However, so far, there was no possibility of reproduction the apex of the heart point. The paper presents the nature of the problem and suggest solution. The results of simulation tests and analysis of assumptions correctness are shown. Description of further development works is included.
Recent advances in latent space dynamics model from pixels show promising progress in vision-based model predictive control (MPC). However, executing MPC in real time can be challenging due to its intensive computatio...
详细信息
ISBN:
(数字)9781665490429
ISBN:
(纸本)9781665490429
Recent advances in latent space dynamics model from pixels show promising progress in vision-based model predictive control (MPC). However, executing MPC in real time can be challenging due to its intensive computational cost in each timestep. We propose to introduce additional learning objectives to enforce that the learned latent space is proportional derivative controllable. In execution time, the simple PD-controller can be applied directly to the latent space encoded from pixels, to produce simple and effective control to systems with visual observations. We show that our method outperforms baseline methods to produce robust goal reaching and trajectory tracking in various environments.
A kinematic method is developed to analyze the workholding condition by evaluating the motion stops corresponding to the reciprocal screw motions within a given fixture configuration. More significantly, the method ca...
详细信息
ISBN:
(纸本)0818620617
A kinematic method is developed to analyze the workholding condition by evaluating the motion stops corresponding to the reciprocal screw motions within a given fixture configuration. More significantly, the method can be used to compare the relative quality of two or more configurations in terms of the overall kinematic constraint. Graphically based methods are then developed which can be used to synthesize a fixture layout configuration for a given 3-D workpart geometric model. A CAD system is used to demonstrate the techniques for automated fixture layout planning. The results of this work have been applied directly to a set of modular fixture elements for sheet metal workparts. For simple geometries, the fixture configurations chosen with the motion stop method agree well with the intuitive choices an engineer would make. However, for complicated geometries, these methods provide analysis and synthesis solutions which would not otherwise be possible.
Constrained minimum variance control is offered for nonsquare LTI MIMO systems. A constrained control design takes advantage of the so-called control zeros. The new control strategy is compared with familiar generaliz...
详细信息
Constrained minimum variance control is offered for nonsquare LTI MIMO systems. A constrained control design takes advantage of the so-called control zeros. The new control strategy is compared with familiar generalized minimum variance control and possible application areas of the two are discussed.
One of the key challenges in using reinforcement learning in robotics is the need for models that capture natural world structure. There are methods that formalize multi-object dynamics using relational representation...
详细信息
Heating, ventilation, and air conditioning (HVAC) systems in commercial buildings consume a large proportion of energy in the world. One possible way to reduce the cost is to optimize HVAC operation through predictive...
详细信息
ISBN:
(纸本)9781479928378
Heating, ventilation, and air conditioning (HVAC) systems in commercial buildings consume a large proportion of energy in the world. One possible way to reduce the cost is to optimize HVAC operation through predictive control methods. Such approaches rely on dynamic models to meet comfort constraints and to minimize energy usage. In this paper, we employ two different model types: an resistance-capacitance (R-C) network model and an artificial neural network (ANN) model, in order to model thermal dynamics of an airport terminal building. The R-C model serves as a control model and its parameters are identified using a gray box identification technique. The ANN model is built in recursive form and works as a simulator to provide system feedback to the controller. We implement a model predictive control (MPC) with two supervisory scenarios on such a simulation environment to evaluate their energy-saving potential. Simulation results during cooling season show 5% to 18% of daily energy savings can be achieved when the proposed MPC is applied to the building.
This study presents the development of a novel calculation model for the influence function method, specifically tailored for the complete roll system of a 20-high rolling mill, a critical component in strip shape con...
详细信息
Localization is the problem of determining a robot's location in an environment. Monte Carlo Localization (MCL) is a method of solving this problem by using a partially observable Markov decision process to find t...
详细信息
ISBN:
(纸本)9728865600
Localization is the problem of determining a robot's location in an environment. Monte Carlo Localization (MCL) is a method of solving this problem by using a partially observable Markov decision process to find the robot's state based on its sensor readings, given a static map of the environment. MCL requires a model of each sensor in order to work properly. One of the most important sensors involved is the estimation of the robot's motion, based on its encoders that report what motion the robot has performed. Since these encoders are inaccurate, MCL involves using other sensors to correct the robot's location. Usually, a motion model is created that predicts the robot's actual motion, given a reported motion. The parameters of this model must be determined manually using exhaustive tests. Although an accurate motion model can be determined in advance, a single model cannot optimally represent a robot's motion in all cases. With a terrestrial robot the ground surface, slope, motor wear, and possibly tire inflation level will all alter the characteristics of the motion model. Thus, it is necessary to have a generalized model with enough error to compensate for all possible situations. However, if the localization algorithm is working properly, the result is a series of predicted motions, together with the corrections determined by the algorithm that alter the motions to the correct location. In this case, we demonstrate a technique to process these motions and corrections and dynamically determine revised motion parameters that more accurately reflect the robot's motion. We also link these parameters to different locations so that area dependent conditions, such as surface changes, can be taken into account. These parameters might even be used to identify surface changes by examining the various parameters. By using the fact that MCL is working, we have improved the algorithm to adapt to changing conditions so as to handle even more complex situations.
This paper proposes a computationally efficient approach to detecting objects natively in 3D point clouds using convolutional neural networks (CNNs). In particular, this is achieved by leveraging a feature-centric vot...
详细信息
The Zhang neural network (ZNN), as a special class of recurrent neural network (RNN), has been proposed by Zhang et al. for the online solution of various time-varying problems. More importantly, such a ZNN is based o...
详细信息
ISBN:
(纸本)9781467347082;9781467347075
The Zhang neural network (ZNN), as a special class of recurrent neural network (RNN), has been proposed by Zhang et al. for the online solution of various time-varying problems. More importantly, such a ZNN is based on the Zhang function (ZF) as the error-monitoring function, which is indefinite and quite different from the usual error functions in the study of conventional algorithms, such as a scalar-valued norm-based energy function involved in the gradient-based neural network (GNN). Meanwhile, the resultant ZNN model can guarantee the global/exponential convergence performance for online time-varying problems solving by following Zhang et al.'s design method. In this paper, focusing on solving the time-varying complex matrix-inversion problem, the complex ZNN models are proposed, developed and investigated for time-varying complex matrix inversion. In addition, by introducing different complex ZFs, different corresponding complex ZNN models can be proposed and developed for time-varying complex matrix inversion. Finally, through some simulations and verifications, the illustrative results substantiate the efficacy of the complex ZNN models based on different complex ZFs for time-varying complex matrix inversion.
暂无评论