Skynet meets the Swarm: how the Berkeley Overmind won the 2010 StarCraft AI competitionWe’re gathered in a conference room on the Berkeley campus, the detritus of a LAN party scattered around us. The table is covered with computers and pizza, and there’s a game of StarCraft projected on the screen. Oriol Vinyals, a PhD student in computer science, is commanding the Terran army in a life-or-death battle against the forces of the Zerg Swarm.Oriol is very good—one-time World Cyber Games competitor, number 1 in Spain, top 16 in Europe good. But his situation now is precarious: his goliath walkers are holding off the Zerg’s flying mutalisks, but they can’t be everywhere at once. The Zerg player is crafty, retreating in the face of superior firepower but never going far, picking off targets of opportunity and applying constant pressure.Then Oriol makes a mistake. He moves his goliaths slightly out of position, just for a few seconds. It’s enough. The mutalisks react instantly, streaming through the gap in his defenses and making straight for his vulnerable workers. By the time Oriol brings the goliaths back to drive off the mutalisks, his workers are wiped out and his resource production is crippled. Oriol makes a desperate, last-ditch attack on the Zerg base, trying to break through before the mutalisks are reinforced, but it’s too late. One after another, his goliaths get ripped apart by the Zerg defenses. As a new wave of mutalisks emerges from the Zerg hatcheries, he has no choice but to concede—to the computerized AI that just defeated him......The StarCraft AI competition was created to harness and promote StarCraft as a research environment. AI researchers have used RTS games in the past, but their efforts were hampered by the technology available. Open-source games were buggy and untested, and commercial games like StarCraft were inaccessible. This changed in early 2009 with the release of the Brood War API (BWAPI), an open-source toolkit developed by a group of enthusiasts that gives direct access to the game. Ben Weber, a student in the Expressive Intelligence Studio at UC Santa Cruz, had been working on RTS game-based research. He realized that StarCraft and BWAPI could make an immediate impact on his work and be a valuable tool for the AI community. He set about organizing a tournament for StarCraft AI agents to compete against each other, hoping to kick-start progress and raise interest. The announcement for the tournament was made in November of 2009, and the word soon went out on gaming websites and blogs: the 2010 Artificial Intelligence and Interactive Digital Entertainment (AIIDE) Conference, to be held in October 2010 at Stanford University, would host the first ever StarCraft AI competition......In theory, a computer should be great at controlling many units simultaneously, since it’s not limited by human speeds. Indeed, there is a common misconception that because StarCraft is real-time, it must be game of reflexes. But while speed is useful and important, it is no substitute for knowing the right thing to do. Good unit management requires intelligent decision-making. Simply moving the mutalisks in combat demands some sophistication. The mutalisks need to fly toward their targets but stay away from other enemies; they need to concentrate their firepower, yet disperse when facing enemy units with area attacks. To top it off, dozens of mutalisks must do this in sync.To handle these issues and limit computational overhead, our agent uses artificial potential fields for unit movement. The potential field controller generates virtual forces that push the mutalisks around, balancing attractive forces on targets with repulsive forces on threats. Summing up the forces acting on a mutalisk gives a direction to fly, resulting in a simple but robust control scheme. We were helped by the fact that air units like mutalisks don’t get stuck on obstacles or each other.Potential field control is very powerful. For example, hit-and-run maneuvers can be created by simply turning off attractive target potentials after firing; repulsive threat potentials automatically cause the mutalisks to dance out of range while waiting for their attacks to recharge. Attractive potentials draw the group together to concentrate fire, but if an area-attack threat appears, a repulsive potential causes the cluster to quickly scatter. The video bellow shows this behavior in an engagement with Protoss Archons, powerful anti-air units. With potential fields, “you get certain desired behaviors that just emerge,” says David Burkett. ......For the Zerg, much mid and late-game scouting information comes from proper use of overlords. Overlords are slow-moving flying units with long sight-range that also provide supply, an in-game metric governing the number of units that can be built. Having overlords scattered around the map improves targeting and macro planning, but losing overlords both reduces vision of the map and hinders the ability to build more units. The benefit gained from observations has to be balanced against the risk of losing overlords, and the overlords need to be protected from blundering into enemy units.Our solution to the problem of overlord control and scouting had an uninspired beginning. StarCraft’s built-in path planning for ground units is terrible, an irritant that has hindered players for over a decade. As development progressed, Dan decided that we weren’t going to put up with the indignity of watching units getting stuck on walls and chasing their own tails, so we implemented our own path planning.Simply getting from point A to point B successfully was useful enough, but the real change came from combining path planning with an awareness of enemy units. The Overmind keeps a continuously updated map of the positions of every enemy unit it has ever seen and the last known location of that unit. Since the attack range and speed of units in StarCraft are known, the agent can combine this information with its map of enemy units to build a threat map. For each unit type, it can calculate a level of danger for that unit to be in an area. This threat map can be combined with the path planning algorithm by including the threat in a given map area as part of the cost of traversing it. Short paths that are under high threat are less preferable to longer, safer paths. Modifying the algorithm this way required some technical tricks to run quickly enough to be useful, but once working it proved to be a valuable tool. Though originally intended for worker units, we were suddenly using it everywhere........
Wham alyu men dead awah? i dont see u guys online anymore maybe some select few...add meh those who still play
lol, blame mass effect 2.Then reach fever, then blackops fever, then vietnam and so on and so forth.too many good games, too little time be the cry.Just fired up War for Cybertron today and it bess. Autobots, tranform.Not sure when I'll get back to my beloved SC2 at this rate.Need like 3 months vacation.
Quote from: TriniWyatt on February 06, 2011, 10:31:34 PMlol, blame mass effect 2.Then reach fever, then blackops fever, then vietnam and so on and so forth.too many good games, too little time be the cry.Just fired up War for Cybertron today and it bess. Autobots, tranform.Not sure when I'll get back to my beloved SC2 at this rate.Need like 3 months vacation.aye you playin that legit on steam??
Sunday it is, we deh from 7 pm Trinitime
