Probabilistic Semantic Data Association for Collaborative Human-Robot Sensing

TL;DR

This paper introduces a probabilistic method for semantic data association in human-robot sensing, enabling robots to better interpret and fuse uncertain and ambiguous semantic data from humans for improved collaborative state estimation.

Contribution

It develops the first rigorous probabilistic semantic data association algorithm and integrates it into a Bayesian fusion scheme for robust multi-object search tasks.

Findings

PSDA improves robustness in state estimation with erroneous semantic data

The approach handles significant reference ambiguities effectively

Simulation results demonstrate enhanced collaborative sensing performance

Abstract

Humans cannot always be treated as oracles for collaborative sensing. Robots thus need to maintain beliefs over unknown world states when receiving semantic data from humans, as well as account for possible discrepancies between human-provided data and these beliefs. To this end, this paper introduces the problem of semantic data association (SDA) in relation to conventional data association problems for sensor fusion. It then develops a novel probabilistic semantic data association (PSDA) algorithm to rigorously address SDA in general settings, unlike previous work on semantic data fusion which developed heuristic techniques for specific settings. PSDA is further incorporated into a recursive hybrid Bayesian data fusion scheme which uses Gaussian mixture priors for object states and softmax functions for semantic human sensor data likelihoods. Simulations of a multi-object search task show that PSDA enables robust collaborative state estimation under a wide range of conditions where semantic human sensor data can be erroneous or contain significant reference ambiguities.