Having read Suarez' "Kill Decision" last night, which talks about using computer models of weaver ants for control of autonomous killer drones... wow, coincidence.
This is interesting, but designing the distributed planner is really hard. I think that'll likely be an offline machine learning task - fitting an algorithm that will handle the variances in the construction in a no communication, totally distributed fashion.
This is really awesome. Actually, is the missing piece of my startup, Engarte. When we have these robots, you will be able to hire engineering project and instantly have it built by our army! :)
Comments
Having read Suarez' "Kill Decision" last night, which talks about using computer models of weaver ants for control of autonomous killer drones... wow, coincidence.
This is interesting, but designing the distributed planner is really hard. I think that'll likely be an offline machine learning task - fitting an algorithm that will handle the variances in the construction in a no communication, totally distributed fashion.
This is really awesome. Actually, is the missing piece of my startup, Engarte. When we have these robots, you will be able to hire engineering project and instantly have it built by our army! :)