The Robot World Cup Initiative (RoboCup) is an effort to promote AI and intelligent robotics research by providing a standard problem that integrates various technologies. The first competition will be held at IJCAI-97 in Nagoya. RoboCup involves tasks for a team of multiple fast-moving robots in a dynamic environment, requiring technologies such as autonomous agents, multi-agent collaboration, strategy acquisition, real-time reasoning, robotics, and sensor-fusion. While the ultimate goal is a real-world robot competition, RoboCup also offers a software platform for research on its software aspects. The paper discusses technical challenges, rules, and simulation environments for RoboCup. RoboCup is proposed as a new standard AI problem, focusing on soccer as a platform for AI and robotics research. It aims to provide a standard task for research on fast-moving multiple robots. RoboCup includes three competitions: real robot, software robot, and special skills. Standard AI problems have been fundamental for AI research, but real-world problems are complex and constrained. RoboCup is designed to address these challenges by providing a realistic yet affordable problem for research. The paper discusses research issues for real robots in RoboCup, including the design of players and their control, vision and sensor fusion, and learning behaviors. RoboCup players must perform multiple subtasks, requiring compact and powerful actuators and sophisticated control techniques. Vision and sensor fusion are essential for real-time decision-making, and multi-sensor integration is necessary for accurate information processing. Learning behaviors is crucial for adapting to dynamic environments, with reinforcement learning being a promising approach. Multi-agent collaboration is a key aspect of RoboCup, as soccer is a complex, dynamic environment requiring cooperation among agents. Challenges include dynamic environments, limited perception, varying roles, and communication constraints. RoboCup provides a cooperative distributed planning scheme for common goals, requiring local and global planning coordination. The RoboCup rules for real robots include field size, robot size, goals, ball, and specific regulations for defense zones, robot marking, and fouls. The simulation track uses a 2D soccer field with specific sizes and rules. The simulation environment, Soccer Server, allows clients to control players via UDP/IP, enabling the testing of multi-agent systems. The visualization system provides 3D animation for game observation. RoboCup aims to promote AI and robotics research by providing a standard problem, fostering collaboration, and hosting competitions. The initiative invites participation to define rules, develop research environments, and host events. RoboCup is seen as a significant platform for advancing AI and robotics research.The Robot World Cup Initiative (RoboCup) is an effort to promote AI and intelligent robotics research by providing a standard problem that integrates various technologies. The first competition will be held at IJCAI-97 in Nagoya. RoboCup involves tasks for a team of multiple fast-moving robots in a dynamic environment, requiring technologies such as autonomous agents, multi-agent collaboration, strategy acquisition, real-time reasoning, robotics, and sensor-fusion. While the ultimate goal is a real-world robot competition, RoboCup also offers a software platform for research on its software aspects. The paper discusses technical challenges, rules, and simulation environments for RoboCup. RoboCup is proposed as a new standard AI problem, focusing on soccer as a platform for AI and robotics research. It aims to provide a standard task for research on fast-moving multiple robots. RoboCup includes three competitions: real robot, software robot, and special skills. Standard AI problems have been fundamental for AI research, but real-world problems are complex and constrained. RoboCup is designed to address these challenges by providing a realistic yet affordable problem for research. The paper discusses research issues for real robots in RoboCup, including the design of players and their control, vision and sensor fusion, and learning behaviors. RoboCup players must perform multiple subtasks, requiring compact and powerful actuators and sophisticated control techniques. Vision and sensor fusion are essential for real-time decision-making, and multi-sensor integration is necessary for accurate information processing. Learning behaviors is crucial for adapting to dynamic environments, with reinforcement learning being a promising approach. Multi-agent collaboration is a key aspect of RoboCup, as soccer is a complex, dynamic environment requiring cooperation among agents. Challenges include dynamic environments, limited perception, varying roles, and communication constraints. RoboCup provides a cooperative distributed planning scheme for common goals, requiring local and global planning coordination. The RoboCup rules for real robots include field size, robot size, goals, ball, and specific regulations for defense zones, robot marking, and fouls. The simulation track uses a 2D soccer field with specific sizes and rules. The simulation environment, Soccer Server, allows clients to control players via UDP/IP, enabling the testing of multi-agent systems. The visualization system provides 3D animation for game observation. RoboCup aims to promote AI and robotics research by providing a standard problem, fostering collaboration, and hosting competitions. The initiative invites participation to define rules, develop research environments, and host events. RoboCup is seen as a significant platform for advancing AI and robotics research.