Swarm Localization and Control via On-board Sensing and Computation
Python Files For Gazebo Drone Simulation-20190728T071323Z-001.zip
MATLAB Analysis Of ROS Bags-20190728T083949Z-002.zip
AuthorsRajab, Fat-Hy Omar
AdvisorsShamma, Jeff S.
Permanent link to this recordhttp://hdl.handle.net/10754/656197
MetadataShow full item record
AbstractMulti-agent robotic system have been proved to be more superior in undertaking functionalities, arduous or even impossible when performed by single agents. The increased e ciency in multi agent systems is achieved by the execution of the task in cooperative manner. But to achieve cooperation in multi agent systems, a good localization system is an important prerequisite. Currently, most of the multi-agent system rely on the use of the GPS to provide global positioning information which su ers great deterioration in performance in indoor applications, and also all to all communication between the agents will be required which is not e cient especially when the number of agents is large. In this regard, a real-time localization scheme is introduced which makes use of the robot's on-board sensors and computational capabilities to determine the states of other agents in the multi agent system. This algorithm also takes the advantage of the swarming behaviour of the robots in the estimation of the states. This localization algorithm was found to produce more accurate agent state estimates as compared to a similar localization algorithm that does not take into account the swarming behaviour of the agents in simulations and real experiment involving two Unmanned Aerial Vehicles.
Showing items related by title, author, creator and subject.
A Morphing framework to couple non-local and local anisotropic continuaAzdoud, Yan; Han, Fei; Lubineau, Gilles (International Journal of Solids and Structures, Elsevier BV, 2013-05) [Article]In this article, we develop a method to couple anisotropic local continua with anisotropic non-local continua with central long-range forces. First, we describe anisotropic non-local models based on spherical harmonic descriptions. We then derive compatible classic continuum models. Finally, we apply the morphing method to these anisotropic non-local models and present three-dimensional numerical examples to validate the efficiency of the technique. © 2013 Elsevier Ltd. All rights reserved.
A morphing strategy to couple non-local to local continuum mechanicsLubineau, Gilles; Azdoud, Yan; Han, Fei; Rey, Christian C.; Askari, Abe H. (Journal of the Mechanics and Physics of Solids, Elsevier BV, 2012-06) [Article]A method for coupling non-local continuum models with long-range central forces to local continuum models is proposed. First, a single unified model that encompasses both local and non-local continuum representations is introduced. This model can be purely non-local, purely local or a hybrid depending on the constitutive parameters. Then, the coupling between the non-local and local descriptions is performed through a transition (morphing) affecting only the constitutive parameters. An important feature is the definition of the morphing functions, which relies on energy equivalence. This approach is useful in large-scale modeling of materials that exhibit strong non-local effects. The computational cost can be reduced while maintaining a reasonable level of accuracy. Efficiency, robustness and basic properties of the approach are discussed using one- and two-dimensional examples. © 2012 Elsevier Ltd.
From Local SGD to Local Fixed Point Methods for Federated LearningMalinovsky, Grigory; Kovalev, Dmitry; Gasanov, Elnur; Condat, Laurent; Richtarik, Peter (arXiv, 2020-04-03) [Preprint]Most algorithms for solving optimization problems or finding saddle points of convex-concave functions are fixed point algorithms. In this work we consider the generic problem of finding a fixed point of an average of operators, or an approximation thereof, in a distributed setting. Our work is motivated by the needs of federated learning. In this context, each local operator models the computations done locally on a mobile device. We investigate two strategies to achieve such a consensus: one based on a fixed number of local steps, and the other based on randomized computations. In both cases, the goal is to limit communication of the locally-computed variables, which is often the bottleneck in distributed frameworks. We perform convergence analysis of both methods and conduct a number of experiments highlighting the benefits of our approach.