Update: April 3, 2017

This video series is documenting my thought processes and experimental approaches to an "environment-responsive animation-creating rig". The rig is created in Houdini, but the basic ideas easily transfer to any rigging system that allows for scripted "nodes" (or elements). Obviously the results, including the final version, are not "perfect", there is room and demand for improvements, that I want to encourage viewers to dabble with. The idea is to show that errors, failed experiments and dead ends can spawn creative alternatives, that more often than not turn out to be better than the original plan. This first episode covers scribbling the basic idea, gathering and understanding the required tools (in Houdini), setting up an experimental basic rig and tweaking it until the fundamental (automatic) function can be seen, used and improved on.
This second part of my "documentary series" on the creation of an environment-aware automatic animation rig goes into the depth of me constantly making the same kinds of mistakes over and over again and slowly learning that Houdini loves lots of nodes and dislikes recursive setups ... On the productive side: The rig is getting to a point where is reacts to an arbitrary terrain ground using an "intersect" raycast, moves the "leg holding" body node along a path, also reacting to the ground by the same raycast "groundsensor". A ramp to shape the animation behavior gets added and some enhancements to thresholding and target-finding are made.