In the paper we recover the static characteristic of Wiener-Hammerstein (sandwich) system from inputoutput data. the system is excited and disturbed by random processes with arbitrary distribution. Two kernel-based es...
详细信息
this paper concerns two fundamental issues in video shots retrieval: key frame identification and similarity measurement between the key frames. We propose a simple key frame extraction algorithm based on optical flow...
详细信息
the contribution describes an application of the algebraic μ-synthesis to the control of a real plant with a nonlinear characteristic. the controller design is considered as a problem of minimization of the peak of t...
详细信息
ISBN:
(纸本)9781629934839
the contribution describes an application of the algebraic μ-synthesis to the control of a real plant with a nonlinear characteristic. the controller design is considered as a problem of minimization of the peak of the structured singular value denoted μ. the algebraic approach consists of the pole placement principle based on the polynomial Diophantine equations and Differential Migration procedure used for optimization. the results are compared with other controllers designed via the D-K iteration, synthesis in the ring of proper and stable functions and the Naslin method.
An approach for nonholonomic two-wheeled mobile robot trajectory tracking and obstacle avoiding is presented in this paper. If the desired trajectory is provided by high level planner, trajectory tracking problem can ...
详细信息
ISBN:
(纸本)9789898565211
An approach for nonholonomic two-wheeled mobile robot trajectory tracking and obstacle avoiding is presented in this paper. If the desired trajectory is provided by high level planner, trajectory tracking problem can be solved in various ways. In this paper, tracking is provided using proportional-integral (PI) or fuzzy logic controller (FLC). Unfortunately, tracking is never perfect, due to uncertainties and obstacles can change their positions in time. In order to overcome these difficulties, additional correction controller must be used. Here is proposed fuzzy controller, which slightly changes control action of the tracking controller in order to prevent collision with obstacles. this approach is proved to be efficient even in dynamic environments. Simulation results are presented as illustration of the proposed approach.
Refined algorithms for solving continuous-time algebraic Riccati equations using Newton's method with or without line search are discussed. their main properties are briefly presented. Algorithmic details incorpor...
详细信息
ISBN:
(纸本)9789897580390
Refined algorithms for solving continuous-time algebraic Riccati equations using Newton's method with or without line search are discussed. their main properties are briefly presented. Algorithmic details incorporated in the developed solver are described. the results of an extensive performance investigation on a large collection of examples are summarized. Several numerical difficulties and observed unexpected behavior are reported. these algorithms are strongly recommended for improving the solutions computed by other solvers.
Deep Learning experiments commonly require hundreds of trials to properly train neural networks, often labeled as Big Data, while Bayesian learning leverages scarce data points to infer next iterations, also known as ...
详细信息
ISBN:
(纸本)9789897585227
Deep Learning experiments commonly require hundreds of trials to properly train neural networks, often labeled as Big Data, while Bayesian learning leverages scarce data points to infer next iterations, also known as Micro Data. Deep Bayesian Learning combines the complexity from multi-layered neural networks to probabilistic inferences, and it allows a robot to learn good policies within few trials in the real world. In here we propose, for the first time, an application of Deep Bayesian Reinforcement Learning (RL) on a realworld multi-robot confrontation game, and compare the algorithm with a model-free Deep RL algorithm, Deep Q-Learning. Our experiments show that DBRL significantly outperforms DRL in learning efficiency and scalability. the results of this work point to the advantages of Deep Bayesian approaches in bypassing the Reality Gap and sim-to-real implementations, as the time taken for real-world learning can quickly outperform data-intensive Deep alternatives.
As an engine of almost all. life phenomena, the motor information generated by the central nervous system (CNS) plays a critical role in the activities of all animals. Despite the difficulty of being physically identi...
详细信息
ISBN:
(纸本)9789898111302
As an engine of almost all. life phenomena, the motor information generated by the central nervous system (CNS) plays a critical role in the activities of all animals. Despite the difficulty of being physically identified, the central pattern generator (CPG), which is a concrete branch of studies on the CNS, is widely recognised to be responsible for generating rhythmic patterns. this paper presents a novel, macroscopic and model-independent approach to the retrieval of different patterns of coupled neural oscillations observed in biological CPGs during the control of legged locomotion. Based on the simple graph dynamics, various types of oscillatory building blocks (OBB) can be reconfigured for the production of complicated rhythmic patterns. Our quadrupedal locomotion experiments show that an OBB-based artificial CPG model alone can integrate all gait patterns and undergo self-organised gait transition between different patterns.
As robotic technology advances, robots become increasingly ubiquitous in the lives of humans. To facilitate this ubiquity, systems must be created that allow robots to interact easily with human beings. these systems ...
详细信息
ISBN:
(纸本)9781479964666
As robotic technology advances, robots become increasingly ubiquitous in the lives of humans. To facilitate this ubiquity, systems must be created that allow robots to interact easily with human beings. these systems must provide for communication between an arbitrary number of agents;for example, a home automation system with half a dozen environment sensors or a city-wide fire prevention system with hundreds of firefighting robots. In addition, these systems must support agents of varying complexity levels which could range from simple temperature sensors to database systems to autonomous robots. In this paper, we present the results of an experimental implementation of a HARMS-based system, the model for which has been proposed by Lewis et al [1].
An overview is given on the integration of product modeling with robot control. A method to enhance application of geometric and other product and production equipment related information in robot control is introduce...
详细信息
An overview is given on the integration of product modeling with robot control. A method to enhance application of geometric and other product and production equipment related information in robot control is introduced. Focus is on creation and application of model data information understandable by robot controls, on communication of model data with robot environments through the Internet, and finally on some related telerobotics issues.
Passively compliant drives are interesting alternatives to classical stiff actuators in emerging fields like human- robot cooperation, service and rehabilitation robotics. Pneumatic muscles have been found to be inter...
详细信息
ISBN:
(纸本)9789898565211
Passively compliant drives are interesting alternatives to classical stiff actuators in emerging fields like human- robot cooperation, service and rehabilitation robotics. Pneumatic muscles have been found to be interesting low-cost actuators for such purposes. To fully realize the (desired) higher sensitivity and at the same time maintain a good control quality, detailed models of the robot's own components are required. For pneumatic muscles, their hysteresis characteristic is a challenging property. In this paper we present a hysteresis model based on a Prandtl-Ishlinskii operator approach and evaluate the resulting performance when the inverse model is used for compensation in the position controller. the evaluation is done on a real multi-axes robot arm.
暂无评论