Tiny Recursive Control: Iterative Reasoning for Efficient Optimal Control

TL;DR

Tiny Recursive Control (TRC) introduces a compact neural architecture that leverages iterative refinement to achieve near-optimal control with minimal parameters and computational resources, suitable for embedded aerospace systems.

Contribution

TRC demonstrates that control capacity can emerge from iteration depth rather than parameter size, enabling efficient neural control for resource-constrained environments.

Findings

Achieves near-optimal control costs in nonlinear tasks

Requires only millisecond inference on GPU

Uses under 10MB memory, outperforming language models

Abstract

Neural network controllers increasingly demand millions of parameters, and language model approaches push into the billions. For embedded aerospace systems with strict power and latency constraints, this scaling is prohibitive. We present Tiny Recursive Control (TRC), a neural architecture based on a counterintuitive principle: capacity can emerge from iteration depth rather than parameter count. TRC applies compact networks (approximately 1.5M parameters) repeatedly through a two-level hierarchical latent structure, refining control sequences by simulating trajectories and correcting based on tracking error. Because the same weights process every refinement step, adding iterations increases computation without increasing memory. We evaluate TRC on nonlinear control problems including oscillator stabilization and powered descent with fuel constraints. Across these domains, TRC achieves near-optimal control costs while requiring only millisecond-scale inference on GPU and under 10~MB memory, two orders of magnitude smaller than language model baselines. These results demonstrate that recursive reasoning, previously confined to discrete tasks, transfers effectively to continuous control synthesis.