Construction using autonomous agents in a simulated environment

A behavior-based architecture with a connectionist action selection mechanism is introduced that enables a society of autonomous agents to construct arbitrary structures in their simulated two-dimensional world. All objects in the environment are colored discs. Agents can move in their environment and sense discs located around them through distance sensors. An agent can also pick up discs close to them and carry this disc as it moves to be dropped at another location. Construction in this environment involves the agents picking up discs, and then moving to incomplete parts of the structure being built and dropping these discs. The order in which parts of the structure are built affects the completion time of the construction task since discs can obstruct the paths of agents.; An agent has both reactive behaviors which are used primarily to maintain the viability of the agent and navigational planning behaviors that are used for construction. The navigational planning behaviors use an egocentric grid-based representation of the world. Path planning is implemented by spreading activations on sets of grid-based maps. The shape of the structure to be built is also encoded on an internal spatial map.; The connectionist action selection mechanism can learn to exploit spatial and temporal regularities in the environment using its reactive behaviors and also learn to perform sequences of navigational planning behaviors by imitation learning. Each agent monitors its progress to detect deadlocks arising from interactions with other agents and uses unsupervised learning to change its behavior so that the deadlock is broken. Algorithms are implemented on the spatial maps to decide the order in which discs are to be picked up and carried to incomplete parts of the structure in order to reduce the time taken to complete construction. Communication between agents is used to reduce the effect of random sensory and odometry errors on the accuracy of the spatial maps.; We present simulation results that show how the various algorithms perform as parameters and environmental factors are modified.