CS Freiburg: coordinating robots for successful soccer playing

A dynamic role assignment algorithm is proposed in the paper for formation control of multiple mobile robots. The goal of the algorithm is to reassign a role for each robot automatically during a formation is forming or switching. Many formation control systems have been successfully implemented and validated by supporting experimental results. Nevertheless, this research aims at providing an efficient algorithm of role assignment for a class of formation control systems employing the concept of combinational optimization problems. Specifically, by exploring spatial relationship between robots and information of obstacles surrounding the robots, a character cost function is found to represent the degree of difficulty for a robot been assigned a specified role in a formation. Instead of using complex cost minimization procedure, a solution is provided by calculating the largest value of character set fitness, and a new formation is selected for robots accordingly. The developed algorithm is applied to the formation control of a group omni-directional driven robots. Simulation and experimentation are performed with real platform to verify the proposed algorithm and the results show that the performance of the proposed dynamic role assignment algorithm is efficient for robot formation control.

Mobile robots have ability to move their entire body and doing some tasks automatically. One of them is RoboCup Middle Size League (MSL) soccer robot. Main systems in the MSL robots are a self-localization. Localization system is very important because it can help some important aspects such as ability to navigate and control. The common localization method of mobile robots is odometry because the process is fast, but the weakness is that the error from odometry will increase over time due to gyro drifting and encoder wheels are slipping. Meanwhile, MSL robots generally use omnivision with the particle filter method for localization. Localization error with omnivision does not increase over time but requires heavy computational processing. Therefore, a sensor fusion system was designed to combine odometry and omnivision. Thus, they can cover each other’s disadvantages, and make localization become more accurate. From the experimental results on the soccer field with size $9{\mathrm {m}}\times 6{\mathrm {m}}$, sensor fusion can provide good localization data. The localization error results x = 10.5 ± 7.8 cm, y = 7.6 ± 6.8 cm, and $\theta$ = 1.9 ± 1.2°, with average time response 1.6 ms. This system is expected to give soccer robot more accurate localization in Robocup MSL matches and help robot navigation when avoiding obstacles.

Self-localization is the basis to realize autonomous ability such as motion planning and decision-making for mobile robots, and omnidirectional vision is one of the most important sensors for RoboCup Middle Size League (MSL) soccer robots. According to the characteristic that RoboCup competition is highly dynamic and the deficiency of the current self-localization methods, a robust and real-time self-localization algorithm based on omnidirectional vision is proposed for MSL soccer robots. Monte Carlo localization and matching optimization localization, two most popular approaches used in MSL, are combined in our algorithm. The advantages of these two approaches are maintained, while the disadvantages are avoided. A camera parameters auto-adjusting method based on image entropy is also integrated to adapt the output of omnidirectional vision to dynamic lighting conditions. The experimental results show that global localization can be realized effectively while highly accurate localization is achieved in real-time, and robot self-localization is robust to the highly dynamic environment with occlusions and changing lighting conditions.

Contextual information plays an important role in hiding technological complexities from a user's point of view in ubiquitous and pervasive environments. In terms of access control, contextual information can be used to minimize the set up and access control complexities and at the same time to allow users to gain access to resources pervasively without difficulties. RBAC has been used and adopted widely in facilitating an access control mechanism. A variation of RBAC is dynamic role assignment in which users are assigned roles based on their situations. This paper uses semantic contexts in order to determine appropriate roles assignment for an incident management system.

This paper investigates the dynamic role assignment and collision avoidance for the discrete-time system without global information. Based on the traditional formation control protocol, the formation control algorithm with dynamic role assignment is proposed. To make each agent act efficiently, we give an assignment transformation matrix by solving an optimal problem with distributed auction algorithm. Modified formation control algorithm with derivation of an avoidance function is presented, which can guarantee collision-free movement for each agent. Mathematical theory and simulation verify the effectiveness of the proposed algorithms.

Purpose– The purpose of this paper is to design intelligent robots operating in such dynamic environments like the RoboCup Middle-Size League (MSL). In the RoboCup MSL, two teams of five autonomous robots play on an 18- × 12-m field. Equipped with sensors and on-board computers, each robot should be able to perceive the environment, make decision and control itself to play the soccer game autonomously.Design/methodology/approach– This paper presents the design of our soccer robots, participating in RoboCup MSL. The mechanical platform, electrical architecture and software framework are discussed separately. The mechanical platform is designed modularly, so easy maintainability is achieved; the electronic architecture is built on industrial standards using PC-based control technique, which results in high robustness and reliability during the intensive and fierce MSL games; the software is developed upon the open-source Robot Operating System (ROS); thus, the advantages of ROS such as modularity, portability and expansibility are inherited.Findings– Based on this paper and the open-source hardware and software, the MSL robots can be re-developed easily to participate in the RoboCup MSL. The robots can also be used in other research and education fields, especially for multi-robot systems and distributed artificial intelligence. Furthermore, the main designing ideas proposed in the paper, i.e. using a modular mechanical structure, an industrial electronic system and ROS-based software, provide a common solution for designing general intelligent robots.Originality/value– The methodology of the intelligent robot design for highly competitive and dynamic RoboCup MSL environments is proposed.

International audience

Robotic soccer is nowadays a popular research domain in the area of multi-robot systems. RoboCup is an international joint project to promote research in artificial intelligence, robotics and related fields. RoboCup chose soccer as the main problem aiming at innovations to be applied for socially relevant problems. It includes several competition leagues, each one with a specific emphasis, some only at software level, others at both hardware and software, with single or multiple agents, cooperative and competitive. In the context of RoboCup, the Middle Size League (MSL) is one of the most challenging. In this league, each team is composed of up to 5 robots with a maximum size of 50cm× 50cm, 80cm height and a maximumweight of 40Kg, playing in a field of 18m× 12m. The rules of the game are similar to the official FIFA rules, with minor changes required to adapt them for the playing robots CAMBADA, Cooperative Autonomous Mobile roBots with Advanced Distributed Architecture, is the MSL Soccer team from the University of Aveiro. The project started in 2003, coordinated by the Transverse Activity on Intelligent Robotics group of the Institute of Electronic and Telematic Engineering of Aveiro (IEETA). This project involves people working on several areas for building the mechanical structure of the robot, its hardware architecture and controllers (Almeida et al., 2002; Azevedo et al., 2007) and the software development in areas such as image analysis and processing (Caleiro et al., 2007; Cunha et al., 2007; Martins et al., 2008; Neves et al., 2007; 2008), sensor and information fusion (Silva et al., 2008; 2009), reasoning and control (Lau et al., 2008), cooperative sensing approach based on a Real-Time Database (Almeida et al., 2004), communications among robots (Santos et al., 2009; 2007) and the development of an efficient basestation. The main contribution of this chapter is to present the new advances in the areas described above involving the development of an MSL team of soccer robots, taking the example of the CAMBADA team that won the RoboCup 2008 and attained the third place in the last edition of the MSL tournament at RoboCup 2009. CAMBADA also won the last three editions

Ball handling and dribbling are basics soccer techniques. Therefore, handling and dribbling mechanism becomes important component for MSL (Middle Size League) robot soccer. The implementation of active handling and dribbling systems aims to enable the robot to perform dribbling mechanism when maneuvering, while in idle position use handling mechanism. Feedforward and feedback applied as control method in active ball handling and dribbling system. Feedforward control method will give a response to the dribbler wheel based on the motion of robot movement against x , y , and θ axes. Whereas, Feedback control method applied fuzzy inference system by reading the displacement of the arm angle from sensor databased on angle reference of dribbler arm to keep the ball in an ideal position. Implementation geometric calculation for feedforward control toward x , y , and θ axes movement of the robot that it has resulted RMSE value for left motor is 5.75 and for right motor is 7.01. Fuzzy inference system method is used as a feedback control to manipulate the speed and direction of wheel’s rotation as a result ball remains in the ideal position. The value of arm angle displacement and the combination of the two-dribbler arms are used as input for fuzzy inference system. The results of this research show that the application of the combined feedforward and feedback control methods in active handling and dribbling systems can provide robot-dribbling capabilities. Consequently, the robot has a good ability in handling and dribbling mechanism.

Real-time and accurate robot detection and localization is important for the RoboCup Middle size League (MSL) soccer robots. In the current robot detection methods used by most of the teams, the black-color-based information is used to distinguish robots from the environment, which is not robust if the robot changes its makers' color according to the current rule. Considering the good performance of deep learning on the feature extraction and object detection, in this paper, we propose a novel approach for robot detection and localization based on Convolutional Neural Networks (CNNs) for RoboCup MSL soccer robots. The approach is composed of two stages: robot detection using the RGB image, and robot localization using the depth point cloud. The high accuracy and mean average precision (mAP) verify that the proposed method is suitable for robot detection during the MSL competition, which will benifit the following strategy design and obstacle avoidance procedures. The proposed approach can be easily exploited to deal with different objects and adapted to be used in other RoboCup leagues. The acquired dataset is made available for the community.

Large Language Model (LLM)-based multi-agent systems have emerged as a promising paradigm for tackling complex tasks that exceed individual agent capabilities. However, existing approaches often suffer from coordination inefficiencies, a lack of trust mechanisms, and suboptimal role assignment strategies. This paper presents a novel trust-aware coordination framework that enhances multi-agent collaboration through dynamic role assignment and context sharing. Our framework introduces a multi-dimensional trust evaluation mechanism that continuously assesses agent reliability based on performance history, interaction quality, and behavioral consistency. The coordinator leverages these trust scores to dynamically assign roles and orchestrate agent interactions while maintaining a shared context repository for transparent information exchange. We evaluate our framework across eight diverse task scenarios with varying complexity levels, demonstrating significant improvements over baseline approaches. Experimental results show that our trust-aware framework achieves a 87.4% task success rate, reducing execution time by 36.3% compared to non-trust-based methods, while maintaining 43.2% lower communication overhead. The framework's ability to adapt agent roles based on evolving trust scores enables more efficient resource utilization and robust fault tolerance in dynamic multi-agent environments.

Middle-size robot soccer is one of the divisions that competed in national events such as the National Indonesia Robotics Competition and international competitions such as the middle size league (MSL). One of the main components in soccer robots is the kicker system. The kicker system is expected to be high torque, robust, and safe. In this work, a high voltage kicker system is designed and evaluated to substitute ROSTU's previous kicker system. This high voltage solenoid-based kicker system works at 380V and uses the electromagnetic force principle to move a ball. The performance criteria of the kicker system are it can move a ball with a mass of around 1 kg for a minimum range of 3 m and control the charging and discharging process in high voltage conditions. The experiment results show that the kicker system can move a ball with a mass of 1.06 kg, a difference kick distance from 100cm to 350cm, and a monitoring system that can show information about the capacitor voltage and system status.

This paper presents the architecture and key components of a simulated humanoid robot soccer team, UT Austin Villa, which was designed to compete in the RoboCup 3D simulation competition. These key components include (1) an omnidirectional walk engine and associated walk parameter optimization framework, (2) an inverse kinematics based kicking architecture, and (3) a dynamic role assignment and positioning system. UT Austin Villa won the RoboCup 2011 3D simulation competition in convincing fashion by winning all 24 games it played. During the course of the competition the team scored 136 goals while conceding none. We analyze the effect of each component in isolation and show through extensive experiments that the complete team significantly outperforms all the other teams from the competition.

DFuse is an architectural framework for dynamic application-specified data fusion in sensor networks. It bridges an important abstraction gap for developing advanced fusion applications that takes into account the dynamic nature of applications and sensor networks. Elements of the DFuse architecture include a fusion API, a distributed role assignment algorithm that dynamically adapts the placement of the application task graph on the network, and an abstraction migration facility that aids such dynamic role assignment. Experimental evaluations show that the API has low overhead, and simulation results show that the role assignment algorithm significantly increases the network lifetime over static placement.

This paper describes a topological navigation system, based on the description of key-places by a reduced number of parameters that represent images associated to specific locations in configuration space, and the application of the developed system to robotic soccer, through the implementation of the developed algorithms to RoboCup Middle-Size League (MSL) robots, under the scope of the SocRob project (Soccer Robots or Society of Robots). A topological map is associated with a graph, where each node corresponds to a key-place. Using this approach, navigation is reduced to a graph path search. Principal Components Analysis was used to represent key-places from pre-acquired images and to recognize them at navigation time. The method revealed a promising performance navigating between key-places and proved to adapt to different graphs. Furthermore, it leads to a robot programming language based on qualitative descriptions of the target locations in configuration space (e.g., Near Blue Goal with the Goal on its Left). Simulation results of the method application are presented, using a realistic simulator.