Rod Brooks was on my thesis committee. His work was inspirational, but the subsumption architecture was badly flawed and ultimately proved to be a dead end. It's simply not true that "the world is its own best model" because wherever you are you can only see a small part of the world. It works great if all you want to do is avoid obstacles, but if you want to do anything complex and goal-directed, a robot really does have to be able to remember things. The reason he was so influential is because the dominant paradigm at the time, the so-called sense-plan-act model, was just as much of a dead end, and on top of that, produced really boring demos so it was doubly doomed. The right answer is to combine the two approaches with a layer of "glue" that connects them together in a so-called "three-layer" architecture [1]. That's actually the way that autonomous robots work today (AFAIK -- I've been out of the business for a very long time now).
My first job out of school nearly 30 years ago was building commercial research autonomous robots. We used a subsumption arch for real time avoidance and path following linked to a reactive planner implemented in C and xlisp for goal directed actions and task execution. While most modern books on robotics skim the history of this area - there are a lot of good ideas from that period of research that can be cheaply revistited. If you know anyone interested in robotics - get them to read Ron's (lisper) three layer architecture paper - its a classic in terms of summarizing previous approaches and it also that provided a foundation for a lot what now drives today's research.
re: NeilV - Braitenberg "Vehicles" book - enough goodness cannot be said about that book. It seems deceptively simple but its a wonderful place to start thinking about and being inspired to building simple beasties.
I wonder under which paradigm the modern robots of iRobot (the Roomba company, where Brooks works today) are developed. Them being three-layer would be a pretty good indictment of pure-subsumption as an architecture.
I quite appreciated Rodney's paper from 1985 titled "A Robust Layered Control System for a Mobile Robot" [1]. I'm sometimes wondering if mobile robotics would look much different without this paper or if we would arrive to something similar anyway.
I even asked Rodney Brooks whether the results of this paper were applied on some robot platform as a follow up research and received a very interesting reply [2]. I agree that I might have worded it poorly but it's still one of my favourite tweets.
Great guy. Here's an anecdote from that time. He spent a few months on sabbatical at our lab ( https://ai.vub.ac.be/ ) in those days. We were preparing robots for a NATO Advanced Study Institute ( https://www.springer.com/gp/book/9783642796319 ), and I was struggling writing the serial driver for a custom embedded computer for this as the system kept crashing (due to a bug in the dram controller). Anyways, Rod Brooks, offered to help me with the coding. It wasn't needed as the code was not the problem, but I don't know many professors that could and would be prepared to dive in that deep.
P.S. I posted this comment in reply to a Brooks reference a few days back, but might be more on topic here.
I love how he's constantly dubious about autonomous robots news on twitter. Yesterday he criticized the Berkeley delivery https://twitter.com/rodneyabrooks/status/1165736688141787136... for not being autonomous and their short distances. He's also very skeptical about self-driving cars in general.
[1] http://www.flownet.com/gat/papers/tla.pdf