Autonomous Robot Soccer Research Articles

Eligibility traces in an autonomous soccer robot with obstacle avoidance and navigation policy

As the thematic literature of machine learning suggests, reinforcement learning falls between the two methods of supervised learning and unsupervised learning. In this method, the learning agent receives a reward or punishment from the environment according to its action. Therefore, the learning agent interacts with the environment through trial and error and learns to choose the optimal action to achieve its goal. In the meantime, the eligibility traces are considered as one of the main mechanisms of reinforcement learning in receiving delayed rewards. In the use of conventional reinforcement learning methods, when the learning agent achieves a goal, only the value function of the last state-action pair changes, but all the trace states are affected based on the eligibility traces. In other words, the delayed rewards are distributed throughout the trace. Like the ant pheromone effect, this method can increase learning speed empirically to some extent. A soccer robot on the field encounters moving obstacles such as balls, rival robots, and home robots, and fixed obstacles such as gates and flags, so its environment is in a very dynamic state. Therefore, due to the dynamic environment, it is an important issue for autonomous soccer robots to avoid obstacles and should be considered in real play. The main idea of this study is to determine the appropriate traces for the robot to move towards the ball and ultimately to score with the approach of avoiding obstacles in simulating a real soccer match. The desired results obtained from a game played indicate a high level of online experience and decision-making power in the face of new situations.

Verification of Robustness against Noise and Moving Speed in Self-localization Method for Soccer Robot

The main focus of the RoboCup competitions is the game of football (soccer), where the research goals concern cooperative multi-robot and multi-agent systems in dynamic environments. In RoboCup, self-localization techniques are crucial for robots to estimate their own position as well as the position of the goal and other robots before deciding on a strategy. This paper presents a self-localization technique using an omnidirectional camera for an autonomous soccer robot. We propose a self-localization method that uses the information on the white lines of the soccer field to recognize the robot position by optimizing the fitness function using a genetic algorithm. Through experiments we verify the robustness of the proposed method against noise and moving speed.

Real-time Self-localization using Model-based Matching for Autonomous Robot of RoboCup MSL

This paper presents a self-localization technique using an omni-directional camera for an autonomous soccer robot. The position information of the robot is important for strategic behavior and cooperative operation. Therefore, we have proposed the self-localization method which generates the searching space based on a model-based matching with white line information of soccer field, and which recognizes the robot position by optimizing the fitness function using genetic algorithm.

ロボカップ用オープンプラットフォームヒューマノイドロボットの開発

We describe an open platform humanoid robot for the RoboCup humanoid league. The RoboCup is a standard problem for developing intelligent robots. Many researchers are trying to develop such kind of the autonomous soccer robot. However, the humanoid robots are complicated and it is difficult to develop the robot from scratch. Therefore, to gather the technology efficiently, some platform robots have been proposed. In this paper, we proposed a novel open platform robot and discuss on the mechanism. We explain the specifications of the robot and how design and construct the robot. We confirm the robustness of the mechanism through experiments.

Online Learning of Robot Soccer Free Kick Plans Using a Bandit Approach

This paper presents an online learning approach for teams of autonomous soccer robots to select free kick plans. In robot soccer, free kicks present an opportunity to execute plans with relatively controllable initial conditions. However, the effectiveness of each plan is highly dependent on the adversary, and there are few free kicks during each game, making it necessary to learn online from sparse observations. To achieve learning, we first greatly reduce the planning space by framing the problem as a contextual multi-armed bandit problem, in which the actions are a set of pre-computed plans, and the state is the position of the free kick on the field. During execution, we model the reward function for different free kicks using Gaussian Processes, and perform online learning using the Upper Confidence Bound algorithm. Results from a physics-based simulation reveal that the robots are capable of adapting to various different realistic opponents to maximize their expected reward during free kicks.

Selectively Reactive Coordination for a Team of Robot Soccer Champions

CMDragons 2015 is the champion of the RoboCup Small Size League of autonomous robot soccer. The team won all of its six games, scoring a total of 48 goals and conceding 0. This unprecedented dominant performance is the result of various features, but we particularly credit our novel offense multi-robot coordination. This paper thus presents our Selectively Reactive Coordination (SRC) algorithm, consisting of two layers: A coordinated opponent-agnostic layer enables the team to create its own plans, setting the pace of the game in offense. An individual opponent-reactive action selection layer enables the robots to maintain reactivity to different opponents. We demonstrate the effectiveness of our coordination through results from RoboCup 2015, and through controlled experiments using a physics-based simulator and an automated referee.

UMA ABORDAGEM MULTIAGENTES PARA SISTEMA DEFENSIVO EM UM TIME DE FUTEBOL DE ROBÔS BÍPEDES

This article describes the defense module developed for an autonomous robot soccer team. Believing artificial intelligence is capable of overcoming limitations of movement such as slow walking, a new behaviour state was created and added to the decision making logic of our agents: a player marker focused on the opponent with current possession of the ball, defensively supporting the ally currently active. This module was developed during the 2014 edition of RoboCup, where it was initially validated and, later on, automatic tests were performed showing an improvement not only on the team's defense, but also in its overall performance. A few variables, such as ball possession, game concentration zone, among others were measured to estimate the impacts of adopting this module.

Realization of Robot-assisted Laser Scanning Measurement System based on Assembly Platform

The assembly platform of complex structures with a variety of CNC tools can achieve automated clamping, riveting, drilling, seamless calibration docking, and can also realize connecting engineering of complex assembly. In this paper, to solve the issue of digital measurement positioning technology in complex-structure assembly platform, an industrial robot-assisted laser scanning system is designed, and its kinematics model is established. At the same time, the trajectory of the robot rotation scan is planned. Finally, the measurement accuracy of the system we have established proves to meet the ac- tual demands by assembling experiment. In the assembly of complex-structure, digitized measurement positioning technology is an important standard of assembly quality. The technology can be realized by the robots equipped with vision meas- urement system, which is real-time scanning, high measuring accuracy and efficiency. In the 1960s, there were researches of combining visual measurement systems and robot in the world. In 1969, Nilsson developed the first intelligent mobile robot named shakey equipped with a camera, rangefinder, and contact sensors at Stanford SRI Institute, and its task was automatic obstacle avoidance and target motion tracking. In the late 1990s, machine vision measurement technology be- gan to develop rapidly in China. Harbin Institute of Technology achieved a fully autonomous soccer robot navigation with heterogeneous binocular active vision systems, analyzed fusion method of foot- ball robot's navigation data based on the binocular coordination, and proposed the binocular coordina- tion navigation model and target implementation strategy based on its behavior (1). National University of Defense Technology studied surgical instrument tracking and positioning technology based on bin- ocular vision (2). Currently, machine vision technology is widely used in various aspects, especially in industrial production and assembly. The machine vision measurement technology has become increas- ingly mature, but the study of improving the measurement accuracy of vision measurement system is still a main direction that we need to further study. In this paper, through analyzing the complex-structure assembly and detection system, a laser scan- ning measurement system is built by the Swedish ABB robot equipped with a CCD(Charge-coupled Device) high-speed camera. The system proves to be large measurement range, high accuracy and real- time.

Design and Implementation of an Autonomous Robot Soccer System

Abstract A design and implementation method of a robot soccer system with three vision-based autonomous robots is proposed in this paper. A hierarchical architecture with four independent layers: (a) information layer, (b) strategy layer, (c) tactics layer, and (d) execution layer, is proposed to construct a flexible and robust vision-based autonomous robot soccer system efficiently. Five mechanisms, including (a) a two-dimensional neck mechanism, (b) dribbling mechanism, (c) shooting mechanism, (d) aiming mechanism, and (e) flexible movement mechanism, are proposed to mean the robot with multiple functions can win the game. A method based on data obtained from a compass and a vision sensor is proposed to determine the location of the robot on the field. In the strategy design, a hierarchical architecture of decision based on the finite-state transition mechanism for the field players and the goalkeeper is proposed to solve varied situations in the robot soccer game. Three vision-based robots are implemented and some real competition results in the FIRA Cup are presented to illustrate the validity and feasibility of the proposed method in autonomous robot soccer system design.

Design of Autonomous Robot Soccer System Based on Sugeno Fuzzy

This paper presents the use of a fuzzy logic controller combined with Machine Vision to improve the accuracy to identify ball and speed to approaching ball in Robot Soccer. The conventional robot control consists of methods for path generation and path following. When a robot moves away the desired track, it must return immediately, and while doing so, the obstacle avoidance behavior and the effectiveness of such a path are not guaranteed. So, motion control is a difficult task, especially in real time and high speed control. To achieve good control performance and requirements for timeliness, accuracy to identify and approach the ball, the paper designs a sugeno fuzzy control system, which could make robot complete the whole football performance, including finding, approaching and shooting ball. At the end of the paper, to validate the proposed fuzzy algorithm, the experiment on METALFIGHTER-2 humanoid robot has been run, which showed that the fuzzy controller is effectiveness and robustness.

Design and implementation of a fully autonomous humanoid soccer robot

PurposeThe purpose of this paper is to design and implement a team of kid size humanoid soccer robots conforming to RoboCup Humanoid league.Design/methodology/approachThe project is described in two main parts: hardware and software. The hardware section consists of the mechanical structure and the driver circuit board enabling each robot to walk, fast walk, autonomously get up, kick and dribble when it catches the ball. The software is developed as a robot application which consists of motion controller, autonomous motion robot, self localization based on vision system, AI, trajectory planning and network.FindingsThis year, the authors' developments for the humanoid robot include: the design and construction of our new humanoid robots structure and implementation of a new recurrent hybrid neural network for walking control. The control system consists of two neural network controllers, two standard PD controllers and a robot walking planar. The proposed neural network controller has three layers, which are input, hidden, and output layers.Originality/valueThis paper presents results of research work in the field of autonomous robot‐middle size soccer robot, supported by IAU‐ Isfahan Branch (Khorasgan).

It shoots - it scores! [RoboCup

The prime driver for this sporting possibility is the RoboCup international robotics competition. Founded in 1997, its official aim is that, by 2050, a team of fully autonomous humanoid robot soccer players will play against the winner of the most recent World Cup under the rules of football's governing body FIFA -and win.

Design and Realization of the Hybrid Vision System for Autonomous Soccer Robot

A hybrid vision system is designed and implemented according to the need of middle-size league in RoboCup,which is composed of omni-vision,monocular front vision and embedded vision.Omni-vision is of 360° horizontal view angle,large information capacity and good real-time performance,but it can't get the precise nearby image information,for which monocular front vision and embedded vision are introduced.Monocular front vision is mainly used to identify the front object not far away,and it can get specific location information.And the monocular front vision and omni-vision are both processed on upper computer of the robot.To get the nearby image information not in front of robot and not to increase upper computer's burden,this paper uses embedded vision in lower computer to identify the dynamic objects or barriers that are near the robot and not in front of it,and solves the problem of getting nearby image information within a range of 360°.Experiment results show that this vision system can successfully identify distant and nearby objects.

Modeling and implementation of a fully autonomous soccer robot based on omni‐directional vision system

PurposeThe purpose of this paper is to design and implement a team of middle size soccer robots to conform RoboCup middle‐size league.Design/methodology/approachFirst, according to the rules of RoboCup, a middle size soccer robot was designed. The proposed autonomous soccer robot consists of the mechanical platform, motion control module, omni‐directional vision module, front vision module, image processing and recognition module, investigated target object positioning and real coordinate reconstruction, robot path planning, competition strategies, and obstacle avoidance. This soccer robot equips the laptop computer system and interface circuits to make decisions.FindingsIn fact, the omni‐directional vision sensor of the vision system deals with the image processing and positioning for obstacle avoidance and target tracking. The boundary‐following algorithm is applied to find the important features of the field. The sensor data fusion method is utilized in the control system parameters, self‐localization, and world modeling. A vision‐based self‐localization, and the conventional odometry systems are fused for robust self‐localization. The localization algorithm includes filtering, sharing, and integration of the data for different types of objects recognized in the environment.Originality/valueThis paper presents results of research work in the field of autonomous robot‐middle size soccer robot supported by IAU‐Khorasgan Branch (Isfahan).

Challenge: Concept of system life and its application to robotics

Artifact systems created by humans interact with the surrounding natural world and have a large-scale influence on our human lives. The most creditable concept to this critical issue of scientific and technological development seems to be “System Life” that is an innovative competence to be embodied into any artifact system for creating harmony in the world of natural entities and artifact systems interacting with each other. System life is defined as a seamless system of sensing, processing, activating and expressing mechanisms governed by system life information. This paper introduces a design approach of robots possessing system life. First, this paper presents the concept of system life comparing it with the conventional design methodology of intelligent systems. Second, the paper introduces an intelligent control methodology using the cubic neural network that the author developed in order to cope with unpredicted failures. Finally, the paper presents various intelligent robots, a skiing robot, autonomous soccer robots, a game playing robot, as new concrete artifact systems designed using the system life concept.

Action control of soccer robots based on simulated human intelligence

A multi-modal action control approach is proposed for an autonomous soccer robot when the bottom hardware is unchangeable. Different from existing methods, the proposed control approach defines actions with the principle of “perception-planning-action” inspired by human intelligence. Character extraction is used to divide the perception input into different modes. Different control modes are built by combining different control methods for the linear velocity and angular velocity. Based on production rules, the motion control is realized by connecting different perceptions to the corresponding control mode. Simulation and real experiments are conducted with the middle-sized robot Frontier-I, and the proposed method is compared with a proportional-integral-derivative (PID) control method to display its feasibility and performance. The results show that the multi-modal action control method can make robots react rapidly in a dynamic environment.

Development a Real-Time Cooperative Approach for Autonomous Soccer Robots Based on World Model Information

Development a Real-Time Cooperative Approach for Autonomous Soccer Robots Based on World Model Information

Development of an Interactive Augmented Environment and Its Application to Autonomous Learning for Quadruped Robots

This paper describes an interactive experimental environment for autonomous soccer robots, which is a soccer field augmented by utilizing camera input and projector output. This environment, in a sense, plays an intermediate role between simulated environments and real environments. We can simulate some parts of real environments, e.g., real objects such as robots or a ball, and reflect simulated data into the real environments, e.g., to visualize the positions on the field, so as to create a situation that allows easy debugging of robot programs. The significant point compared with analogous work is that virtual objects are touchable in this system owing to projectors. We also show the portable version of our system that does not require ceiling cameras. As an application in the augmented environment, we address the learning of goalie strategies on real quadruped robots in penalty kicks. We make our robots utilize virtual balls in order to perform only quadruped locomotion in real environments, which is quite difficult to simulate accurately. Our robots autonomously learn and acquire more beneficial strategies without human intervention in our augmented environment than those in a fully simulated environment.

Safe Multirobot Navigation Within Dynamics Constraints

This paper introduces a refinement of the classical sense-plan-act objective maximization method for setting agent goals, a real-time randomized path planner, a bounded acceleration motion control system, and a randomized velocity-space search for collision avoidance of multiple moving robotic agents. We have found this approach to work well for dynamic and unpredictable domains requiring real-time response and flexible coordination of multiple agents. First, the approach employs randomized search for objective maximization and motion planning, allowing real-time or any-time performance. Next, a novel cooperative safety algorithm is employed which respects agent dynamics limitations while also preventing collisions with static obstacles or other participating agents. An implementation of our multilayer approach has been tested and validated on real robots, forming the basis for an autonomous robotic soccer team

REUSABLE ARCHITECTURE AND TOOLS FOR TEAMS OF LIGHTWEIGHT HETEROGENEOUS ROBOTS

The software framework RoboFrame has been designed to meet the special requirements for teams of lightweight autonomous heterogeneous robot systems. Due to platform abstraction and modern object oriented design, it allows the reuse of components of common robot control software. It can also efficiently be implemented on new platforms and enables different control architectures for different tasks. For the exemplary application in autonomous robot soccer teams configurable and portable algorithms for vision, world modeling, behavior and motion control have been developed on top of the framework. For debugging, controlling and monitoring, an extendable graphical user interface and a generic simulator package have been implemented around the framework. Based on these instruments, different applications for homogeneous and heterogeneous robot teams can be realized in short time.