Weak/fuzzy list of areas for theoretical exploration

- introduce the component of communication latency in the introductory chapter, connectimg single AV latency components to fleet orchestration
- introduce the idea of extending a generalized solution to a componentized framework for handling multiple dimensions/components of latency, extending single AV latency optimization, to realtime fleet communication latency, and beyond
- discuss potential of an adapter layer to quantum computing systems and future obsolescence of current-day computational systems; specifically micro quantum computers for remote micro-fleet computational efficiency/superiority
- explore theoretical concept: consensus of parallel ensemble components (otherwise used for graceful regression) as means of continuous learning of Super Ego optimization; as fallback systems for achieving resiliency (redundancy of critical systems) in adverse, unpredictable, and adversarial conditions
- relationship and differentiation of Super Ego from the Ego (core behavior decision-making outputs)
- discuss fleet orchestration in high risk adversarial conditions with emphasis on weaknesses/strengths/complexities of falling back to fleet perception for guiding autonomous fleet agents that have impaired perception
- relationship to fleet routing optimization using neural networks (via remote cloud services) with fallback to simple greedy algorithms for route planning, touching on cost optimization via graceful regression to simpler more cost effective, but lower fidelity algorithms
- relate to biological systems and mutual aid
- maybe: relate to concepts of agility in both business (and more broadly any organization that needs to make realtime high-risk decisions): connect to lean, flow, and theory of constraints as high-level mental models guiding the /principles/ of decision making and process management, with emphasis on measuring for change (realtime continuous) and leading indicators of change; provide exploratory discussion of how intelligent autonomous systems can surpass the limitations of logic by falling back to simple principles where pure computational methods of data-driven behavior are ineffective
- demonstrate discretization of symbolic maths for usage with universal approximation networks