danesabo/Obsidian
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Obsidian/Writing/ERLM/research-approach/outline.md
Dane Sabo 8184e48dd3 Auto sync: 2025-09-19 16:33:48 (15 files changed) 
M  .task/backlog.data

M  .task/completed.data

M  .task/pending.data

M  .task/undo.data

M  Writing/ERLM/main.aux

M  Writing/ERLM/main.fdb_latexmk

M  Writing/ERLM/main.fls

M  Writing/ERLM/main.log
2025-09-19 16:33:48 -04:00

1.3 KiB
Raw Permalink Blame History

Okay so here's how things will go:

Integrate V design into the workings here

  1. Requirement identification and translation

    1. Point towards hardens lando thing
    2. we're going to do a nuclear start up sequence

  2. Synthesize requirements into a discrete automata

    1. this makes up our mode switching behavior
    2. There's probably going to be a serious amount of refinement required here
    3. Figure out by the structure of the nodes what the purpose of the mode is
      1. Do all traces leave? Do any traces leave? What does this mean for the FUNCTION of the node?

  3. Build controllers that satisfy each mode requirement

    1. Reachability to ensure valid input and output sets?
    2. We can ensure zeno behavior won't happen by looking at the interface between modes
    3. We should also see based on reachability that a well built controller ONLY can enter the modes as specified by the discrete automata
    4. Contract based methods?

  4. Fuck it man, that's like your provability or whatever man.

What are the critical needs?

  1. We need a way to build some operating procedures into controllers for autonomy
  2. How the hell do we know what the goals of each mode are?
  3. How do we know for sure the continuous dynamics will actually get us there?