eess.SYNov 14, 2026

Enforcing Ellipsoidal Constraints in Neural Networks with a Polynomial Safety Layer

Authors: Rolf Findeisen, Sebastian Hirt, Hoang Hai Nguyen, Grace Li Zhang, Hendrik Alsmeier

Organizations: Control and Cyber-Physical Systems Laboratory, Technical University of Darmstadt, Germany · Hardware for Artificial Intelligence, Technical University of Darmstadt, Germany

Abstract

Often the output of a learned mapping must satisfy prescribed constraints. In the control of autonomous systems, for example, the chosen action must keep the system within a collision-free region. We consider the case in which a proposed output is corrected by projection onto a safety ellipsoid. Computing this projection amounts to solving a quadratically constrained quadratic program (QCQP). We construct a neural-network safety layer by unrolling a feasibility-preserving polynomial penalty iteration. The unrolled structure allows the loss gradient to propagate from the feasible output through the safety layer to the network parameters during training. We show how to choose the layer parameters so that every iterate is feasible and give a computable bound on its distance to the exact projection at any prescribed depth. We illustrate the method by training a LiDAR-based policy to control an unmanned aerial vehicle. A privileged model predictive control expert uses the map and the measured vehicle state to generate the training data. In the simulated narrow mazes, the filtered policy avoided wall contact at every tested stage of training, including at initialization, when every unfiltered flight hit a wall. A representative build of the learned policy and its safety layer for an Arm Cortex-M4F microcontroller occupies about 36 KiB of flash, with a RAM budget below 5 KiB.

Figures & tables

Explore similar work

CardsList
  1. ShardNet: Training Neural Controllers with Hard, Non-Convex Constraints

    Jun 29, 2026Long Kiu Chung, Shreyas KousikNeural Network VerificationSafety-Critical Control

  2. Interleaved Projected Gradient Descent for Safe Imitation Learning

    Oct 5, 2026Shengfan Cao, Francesco BorrelliGradient DescentSafe Imitation Learning

  3. Certified Robust Invariant Polytope Training in Neural Controlled ODEs

    Aug 2, 2024Akash Harapanahalli, Samuel CooganNeural Controlled Differential EquationsRobust Control