IKKA: Inversion Classification via Critical Anomalies for Robust Visual Servoing

TL;DR

IKKA is a novel anomaly weighting framework for robust visual servoing that improves control accuracy under challenging conditions by treating critical anomalies as informative features.

Contribution

The paper introduces IKKA, a topologically motivated anomaly weighting method that enhances visual servoing robustness during distribution shifts and stress scenarios.

Findings

IKKA reduces lateral error by 24% under stress conditions.

IKKA increases control throughput from 20.0 to 24.8 Hz.

Non-parametric analysis shows a large effect size (Cliff's delta = 0.79).

Abstract

We introduce IKKA (Inversion Classification via Critical Anomalies), a topologically motivated weighting framework for robust visual servoing under distribution shift. Unlike conventional outlier handling, IKKA treats maverick points as structurally informative observations: points where small perturbations can induce qualitatively different control responses or class assignments. The method combines local extremality, boundary transversality, and multi-scale persistence into a single anomaly weight, W(x) = E(x) x T(x) x M(x), which modulates control updates near ambiguous decision regions. We instantiate IKKA in a CPU-only embedded visual-servoing pipeline on Raspberry Pi 4 and evaluate it across 230 reproducible runs under nominal and stress conditions. In stress scenarios involving dim illumination and transient occlusion, IKKA reduces the 95th-percentile lateral error by 24% relative to a hybrid baseline (0.124 to 0.094) while increasing throughput from 20.0 to 24.8 Hz. Non-parametric analysis confirms a large effect size (Cliff's delta = 0.79).