Aleatoric Uncertainty from AI-based 6D Object Pose Predictors for Object-relative State Estimation

2025-09-01

AI Summary

This work addresses the challenge of quantifying aleatoric uncertainty in deep learning–based 6D object pose estimators for robotic relative state estimation. We propose a lightweight, plug-and-play uncertainty calibration method: it extends any existing pose predictor with only two independent MLP heads while freezing the backbone for efficient training. The estimated pose uncertainties are incorporated in real time into an Extended Kalman Filter (EKF) to dynamically construct the observation noise covariance matrix. Experiments on both synthetic and real-world datasets demonstrate substantial improvements in relative state estimation accuracy over fixed-covariance baselines, with minimal computational overhead suitable for edge deployment. Our core contribution is the first end-to-end framework for estimating 6D pose aleatoric uncertainty and integrating it into filtering—achieving high compatibility, low intrusiveness, and zero modification to the original model architecture.

Technology Category

• Intelligent Robots: State Estimation
• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Uncertainty Representations

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI

Abstract

Deep Learning (DL) has become essential in various robotics applications due to excelling at processing raw sensory data to extract task specific information from semantic objects. For example, vision-based object-relative navigation relies on a DL-based 6D object pose predictor to provide the relative pose between the object and the robot as measurements to the robot's state estimator. Accurately knowing the uncertainty inherent in such Deep Neural Network (DNN) based measurements is essential for probabilistic state estimators subsequently guiding the robot's tasks. Thus, in this letter, we show that we can extend any existing DL-based object-relative pose predictor for aleatoric uncertainty inference simply by including two multi-layer perceptrons detached from the translational and rotational part of the DL predictor. This allows for efficient training while freezing the existing pre-trained predictor. We then use the inferred 6D pose and its uncertainty as a measurement and corresponding noise covariance matrix in an extended Kalman filter (EKF). Our approach induces minimal computational overhead such that the state estimator can be deployed on edge devices while benefiting from the dynamically inferred measurement uncertainty. This increases the performance of the object-relative state estimation task compared to a fix-covariance approach. We conduct evaluations on synthetic data and real-world data to underline the benefits of aleatoric uncertainty inference for the object-relative state estimation task.

Problem

• Estimating aleatoric uncertainty in AI-based 6D pose predictors
• Improving object-relative state estimation with dynamic uncertainty
• Enabling efficient uncertainty inference for robotic navigation tasks

Innovation

• Adding two MLPs for aleatoric uncertainty inference
• Freezing pre-trained pose predictor for efficient training
• Using EKF with dynamically inferred covariance matrix