scholarly journals Multiattribute Utility Copulas for Multi-objective Coverage Control

2014 ◽  
Vol 5 (1) ◽  
Author(s):  
Christopher G. Valicka ◽  
Richard A. Rekoske ◽  
Dušan M. Stipanovic ◽  
Ali E. Abbas
Keyword(s):  
Multirobot Systems ◽  
Control Law ◽  
Objective Functions ◽  
Multiattribute Utility ◽  
Maximum Function ◽  
Differential Drive ◽  
Coverage Control ◽  
Gradient Based ◽  
Global Coverage ◽  
Nonnegative Definite

AbstractThis paper presents theoretical and experimental results related to the control and coordination of multirobot systems interested in dynamically covering a compact domain while remaining proximal, so as to promote robust inter-robot communications, and while remaining collision free with respect to each other and static obstacles. A design for a novel, gradient-based controller using nonnegative definite objective functions and an overapproximation to the maximum function is presented. By using a multiattribute utility copula to scalarize the multiobjective control problem, a control law is presented that allows for flexible tuning of the tradeofs between objectives. This procedure mitigates the controller’s dependence on objective function parameters and allows for the straightforward integration of a novel global coverage objective. Simulation and experiments demonstrate the controller’s efectiveness in promoting scenarios with collision free trajectories, robust communications, and satisfactory coverage of the entire coverage domain concurrently for a group of differential drive robots.

Research of Subsection Optimal Acceleration Control Law of Aircraft Engines Using Fuzzy Switching Based on a Cloud Model

2011 ◽  
Vol 308-310 ◽  
pp. 1242-1247
Author(s):  
Xiao Jie Qiu ◽  
Jin Quan Huang ◽  
Feng Lu
Keyword(s):  
Control Method ◽  
Cloud Model ◽  
Aircraft Engine ◽  
Effective Control ◽  
Control Law ◽  
Acceleration Process ◽  
Aircraft Engines ◽  
Control Parameters ◽  
Objective Functions ◽  
Switching Controller

Considering the problem of optimal acceleration control of aircraft engine, the subsection optimal acceleration control method using fuzzy switching based on a cloud model is proposed. The acceleration process is divided into three phases, optimal objective functions and constraints are different in each phase. The optimal acceleration control law is obtained using the SQP algorithm with adjusted objective functions and constraints to improve the ability of engine acceleration. Considering the significant jump of control parameters and the exceeding limit of aircraft engine state parameters problem during the shift of different phase on the acceleration process, the method of fuzzy switching controller based on a cloud model is designed. Compared with the normal SQP acceleration control law and the subsection one based on direct switching, the simulation results show that the proposed method not only is the most effective control law but also can ensure the aircraft engine stable.

Design of Acoustic Cloak Using Generative Modeling and Gradient-Based Optimization

2021 ◽  
Vol 263 (3) ◽  
pp. 3511-3522
Author(s):  
Linwei Zhuo ◽  
Feruza Amirkulova
Keyword(s):  
Cross Section ◽  
Scattering Cross Section ◽  
Objective Functions ◽  
Total Scattering ◽  
One Dimensional ◽  
Scattering Cross ◽  
Generative Modeling ◽  
Wide Range ◽  
Total Scattering Cross Section ◽  
Gradient Based

Metamaterials are engineered composites that can achieved electromagnetic and mechanical properties that do not exist in natural materials by rearranging their structures. Due to the complexity of the objective functions, it is difficult to find the globally optimized solutions in metameterial design. This talk outlines a gradient-based optimization with generative networks that can search for the globally optimized cloaking devices over a wide range of parameters. The GLO-Net[1] model was developed originally for one-dimensional nano-photonic metagratings is generalized in this work to design two-dimensional broadband acoustic cloaking devices by perturbing positions of each scatterer in planar configuration of cylindrical scatterers. Such optimized cloaking devices can efficiently suppress the total scattering cross section to the minimum at certain parameters over range of wavenumbers. During training each iteration, a generative model generates a batch of metamaterials and compute the total scattering cross section and its gradients using an in-house built multiple scattering MATLAB solver. To evaluate our approach, we compare our obtained results with fmincon in MATLAB. Reference: [1] Jiaqi Jiang and Jonathan A. Fan. Simulator-based training of generative neural networks for the inverse design of metasurfaces. Nanophotonics, 9(5):1059-1069, nov 2019.

scholarly journals Centroidal Voronoi Partitioning using Virtual Nodes for MultiRobot Coverage

2018 ◽  
Vol 7 (2.21) ◽  
pp. 135 ◽  
Author(s):  
Vishnu G Nair ◽  
Guruprasad K R
Keyword(s):  
Load Distribution ◽  
Control Law ◽  
Uniform Load ◽  
Voronoi Cells ◽  
Voronoi Partition ◽  
Gradient Based ◽  
Simulation Results

This paper addresses the problem of Voronoi partitioning using Centroidal Voronoi configuration for a multi-robotic coverage strategy known as Voronoi Partition based Coverage (VPC) algorithm. In VPC, the area to be covered is divided into Voronoi cells and each robot covers the corresponding cell. We use the concept of Centroidal Voronoi Configuration (CVC) to achieve a more uniform load distribution among the robots in terms of the area covered. Instead of the robots moving physically into the CVC, we introduce a concept of virtual nodes, which are deployed into CVC. Once the Voronoi partition is created based on the virtual nods, the robots cover the corresponding Voronoi cells. A gradient based control law has been used for deployment of the virtual nodes. Simulation results are provided to demonstrate the proposed deployment and partitioning scheme. 

Model Predictive Control of a Differential-Drive Mobile Robot

2018 ◽  
Vol 10 (1) ◽  
pp. 20-41
Author(s):  
Samir Bouzoualegh ◽  
El-Hadi Guechi ◽  
Ridha Kelaiaia
Keyword(s):  
Mathematical Model ◽  
Model Predictive Control ◽  
Mobile Robot ◽  
Predictive Control ◽  
Control Law ◽  
Input Output ◽  
Linearization Technique ◽  
Differential Drive ◽  
Predictive Controller ◽  
Simulation Results

Abstract This paper presents a model predictive control (MPC) for a differential-drive mobile robot (DDMR) based on the dynamic model. The robot’s mathematical model is nonlinear, which is why an input–output linearization technique is used, and, based on the obtained linear model, an MPC was developed. The predictive control law gains were acquired by minimizing a quadratic criterion. In addition, to enable better tuning of the obtained predictive controller gains, torques and settling time graphs were used. To show the efficiency of the proposed approach, some simulation results are provided.

Sign in / Sign up

Export Citation Format

Share Document