We consider the problem of learning to control a plant with non-linear control characteristics and solving the path planning problem at the same time. The solution is based on a path planning model that designates a speed field to be tracked, the speed field being the gradient of the stationary solution of a diffusion process. Diffusion is simulated on an artificial neural network by spreading activation. Interneurons between neighboring discretizing neurons detect the strength of the activity flow and emit control signals to control neurons via modifiable connections. The proposed method may be used for learning redundant control problems. The architecture integrates reactive path planning and continuous motion control in a natural fashion.