Kicking It To Goal State

In 1996, IBM computer scientists developed a chess-playing computer named Deep Blue. Not only did people doubt that it would have the ability to beat a world champion, they found it to be ludicrous that it would be able to beat any average joe. The media was in for a shock when on May 11th, 1997 under regular timed situations, Deep Blue beat the world champion at the time, Garry Kasparov at a game of chess.

IBM's Deep Blue innovation has acted as the springboard for the basis of Robocup, and the notion of soccer-playing robots. Robocup's ultimate goal however is by the year 2050, to develop a team of fully autonomous humanoid robots that can win against the human world soccer champion team. RoboCup is an international research and education initiative. It is an attempt to foster artificial intelligence and intelligent robotics research by providing a standard problem which can allow for the integration and analysis of a broad spectrum of technology. In order to satisfy this problem, Robocup chose soccer as the medium in which to explore and integrate technology into. The various measures that must be taken in order for a robot to play a game of soccer includes: design principles of autonomous machines, multi-agent collaboration allowing for team work and optimizing game time situations, strategy acquisition, real-time reasoning, robotics, and sensory functions and actions. RoboCup is a task for a team of multiple fast-moving robots under a dynamic environment, where the ultimate goal is victory. RoboCup also offers a software platform for research on the software aspects , artificial intelligence and programming of RoboCup Although soccer has been chosen as the medium in which to solve a standard problem where broad-range of efforts will be concentrated and integrated, competition is only a part of RoboCup activity, whereas the priority remains to be developing a team of autonomous humanoids that will have the ability to beat a world champion soccer team.

Autonomous soccer-robots, similar to a human soccer-player, would require the ability to have awareness of their surroundings, make on the fly decisions, and carry out dexterous soccer-skills in order to achieve a common goal of scoring goals and preventing an opposition from scoring on oneself. Considering that the following agent is adapting to a current, changing environment, choosing the most efficient way in order to carry out a function, and performs based on intelligent deductions, the soccer robot is therefore classified as a learning agent. Furthermore, if the desired goal of creating autonomous robots to beat a human team, an intelligent, learning component would serve as quite necessary regarding the notions of strategizing and reacting and counteracting to opposing strategies and schemes. In current soccer robot models physical superiority and fluidity of movement are of greater importance, however, in future cases the strategic aspect of the game would play in as far more crucial to victory. While in present times, the learning element playing within these robots isn't of much significance, when the humanoids would begin to face human competition, the ability to learn and analyze tendencies, weaknesses and strengths of one's opponent would play in greatly towards the humanoid's conclusive goal of victory.

The foundation for robots, aside from an intelligent program and a structural body is the presence of sensors and effectors. " Robots are physical agents that perform tasks by manipulating the physical world. To do so they are equipped with effectors such as legs, wheels, joints, and grippers. (p.901, Artificial Intelligence, a modern Approach.)" Considering the fact that we demand these autonomous machines, to move in every direction, kick, pass, attack and defend, the humanoid soccer bots rely heavily on their effectors. Although robocup consists of different divisions with slightly different models in each division respectively, the robot's effectors range from wheels, to lifelike limbs. For the humanoid player the most important effectors are the legs and feet and for the goalkeeper, both the arms and legs, all of which allow the robot to be dynamically stable. Along with effectors, robots have sensors which allow for inputs resulting in actuators sending signals to the effectors to carry out a specific task. Considering the fact that the soccer-game environment is crucial to the humanoid's movement and actions, sensors play a key role within the success of the robot. Through light, motion and sound sensors, the autonomous soccer humanoid is able to navigate the playing field, avoiding most obstructions and locating both the ball and goal. These humanoid robots, however, make great use of proprioceptive sensors in order to self-inform of current state and for the reason of having many motors which carry out different functions. The humanoids also employ various tactile sensors which are mainly used when in confined areas surrounded by obstructions/ players etc. These tactile sensors play a significant role in the humanoid's functioning because of the fact that the robot is constantly faced with either, contact with teammate and opposition, and contact with the ball. As well as these tactile sensors for close range motions, the robots also contain sonar sensors in order to evaluate current surroundings and familiarizing the robot with the environment. In essence, the main objectives of the sensors are to locate the ball and goal, familiarize with current setting and generating awareness for teammates and opponents. Through both the sensors and effectors, robots, and more notably the soccer-playing humanoids, are able to carry out and achieve specifically tailored functions and goals.

Although not a direct relation, a robot's environment can determine the actions and necessary reactions that it must take. As described in chapter 2 of Artificial Intelligence, a modern approach the soccer-playing humanoid's environment can be characterized as the following: Fully observable, strategic, sequential, dynamic, and multi-agent competitive. In terms of a soccer game, the environment is fully observable due to the fact that the agent's sensors are given access to the complete state of the environment, reacting to both the ball and movements of other agents. Additionally the environment is categorized as strategic