There are conclusions that I would draw from this module, and that I hope you will share with me.
The first one is that positioning information is the result of action and should be calculated applying the appropriate laws of physics (I am working on that).
The second conclusion is that there must be an 'inner' representation of at least some of the positioning information, so that the actor can choose appropriate actions. Walking while upside down is pretty ridiculous. In general, there is a sense of 'down', i.e. of gravity, so that walking can be conditional on perceiving that the feet are down and the head is up. The actor presumably 'knows' whether the feet are on the ground and he can walk, or whether he is floating.
The third conclusion arises from the second, that it is important to integrate perception into action.