Titlebook: RoboCup 2018: Robot World Cup XXII; Dirk Holz,Katie Genter,Oskar von Stryk Conference proceedings 2019 Springer Nature Switzerland AG 2019

倔強(qiáng)不能

End-to-End Deep Imitation Learning: Robot Soccer Case Studynable the development of machine learning methods that can get high dimensional data as an input. In this work, we use imitation learning to teach the robot to dribble the ball to the goal. We use B-Human robot software to collect demonstration data and a deep convolutional network to represent the

paroxysm

Designing Convolutional Neural Networks Using a Genetic Approach for Ball Detectionman. This paper describes the design of a convolutional neural network used for the detection of the black and white ball - one of the key contributions that led to the team’s success. We present a genetic design approach that optimizes network hyperparameters for a cost effective inference on the N

prodrome

入會(huì)

Mimicking an Expert Team Through the Learning of Evaluation Functions from Action SequencesThe aim of this paper is to propose a method that improves the performance of a team by mimicking a stronger one. For this purpose, a neural network is employed to model an expert team’s evaluation function. The neural network is trained by using positive and negative episodes of action sequences th

engrossed

Jetson, Where Is the Ball? Using Neural Networks for Ball Detection at RoboCup 2017n. A patch-based classification approach is used. Two different ConvNet architectures, the Inception v3 network by Google and AlexNet are evaluated in the context of a ROS-based architecture on a robot with a Jetson GPU board. The aim is to allow for an efficient re-training of neural networks in th

個(gè)阿姨勾引你

Basal-Ganglia

Multi-Robot Fast-Paced Coordination with Leader Electionnt our solution for leader election among a team, which is based on the Raft algorithm and tackles two of its limitations. The proposed solution was implemented in a real team of soccer-playing robots and the experimental results are thoroughly presented and discussed.

種屬關(guān)系

Infant

Learning Skills for Small Size League RoboCup skills in the CMDragons’ architecture using a physically realistic simulator. We then show how RL can be leveraged to learn simple skills that can be combined by humans into high level tactics that allow an agent to navigate to a ball, aim and shoot on a goal.

剛開始