Estimating Activity at Multiple Scales using Spatial Abstractions

TL;DR

This paper introduces a hierarchical approach combining topology-based clustering and particle filters to estimate and predict navigation activities at multiple spatial scales, improving accuracy and convergence in dynamic environments.

Contribution

It presents a novel hierarchical algorithm that integrates spatial abstractions with particle filtering for activity estimation across multiple scales.

Findings

Improved normalized error in trajectory prediction.

Faster convergence to true activity classes.

Effective in synthetic and real-world datasets.

Abstract

Autonomous robots operating in dynamic environments must maintain beliefs over a hypothesis space that is rich enough to represent the activities of interest at different scales. This is important both in order to accommodate the availability of evidence at varying degrees of coarseness, such as when interpreting and assimilating natural instructions, but also in order to make subsequent reactive planning more efficient. We present an algorithm that combines a topology-based trajectory clustering procedure that generates hierarchically-structured spatial abstractions with a bank of particle filters at each of these abstraction levels so as to produce probability estimates over an agent's navigation activity that is kept consistent across the hierarchy. We study the performance of the proposed method using a synthetic trajectory dataset in 2D, as well as a dataset taken from AIS-based tracking of ships in an extended harbour area. We show that, in comparison to a baseline which is a particle filter that estimates activity without exploiting such structure, our method achieves a better normalised error in predicting the trajectory as well as better time to convergence to a true class when compared against ground truth.