Self-Crystallizing Behavior Trees: Amortizing Foundation-Model Reasoning into Deterministic Control for Autonomous Drones

Key points

• Classical robot control, Finite State Machines and Behavior Trees (BTs), is deterministic and inspectable but fixed at design time: it fails on situations its designers did not anticipate.1
• Foundation models can synthesize new control logic at runtime, at a cost in latency and non-determinism hard to accept in safety-critical aerial systems.1
• Existing LLM-driven approaches keep the model in the loop permanently and discard what it produces.1
• We implement SCBT on Microsoft Project AirSim with a schema-constrained LLM pipeline (Gemini 3.1 Pro) and a lockstep simulator.1

Sources (1)

• [1]Self-Crystallizing Behavior Trees: Amortizing Foundation-Model Reasoning into Deterministic Control for Autonomous Drones
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 18, 06:34 PM

  We propose Self-Crystallizing Behavior Trees (SCBT), in which the foundation model is a compiler rather than a runtime.

  Classical robot control, Finite State Machines and Behavior Trees (BTs), is deterministic and inspectable but fixed at design time: it fails on situations its designers did not anticipate.

Extractive summary: sentences quoted from the sources.