Enhancing Causal Discovery from Robot Sensor Data in Dynamic Scenarios

TL;DR

This paper introduces an enhanced causal discovery method for robot sensor data that combines PCMCI with transfer entropy-based feature selection, improving accuracy and efficiency in dynamic scenarios.

Contribution

It extends PCMCI with a transfer entropy feature-selection module, enabling more accurate and faster causal modeling from robot sensor time-series data.

Findings

Outperforms previous methods in accuracy

Reduces computational time

Effective on real-world robot data

Abstract

Identifying the main features and learning the causal relationships of a dynamic system from time-series of sensor data are key problems in many real-world robot applications. In this paper, we propose an extension of a state-of-the-art causal discovery method, PCMCI, embedding an additional feature-selection module based on transfer entropy. Starting from a prefixed set of variables, the new algorithm reconstructs the causal model of the observed system by considering only its main features and neglecting those deemed unnecessary for understanding the evolution of the system. We first validate the method on a toy problem and on synthetic data of brain network, for which the ground-truth models are available, and then on a real-world robotics scenario using a large-scale time-series dataset of human trajectories. The experiments demonstrate that our solution outperforms the previous state-of-the-art technique in terms of accuracy and computational efficiency, allowing better and faster causal discovery of meaningful models from robot sensor data.