Particle-Based Score Estimation for State Space Model Learning in Autonomous Driving

TL;DR

This paper introduces a particle-based score estimation method for learning parameters in state space models, improving stability and accuracy in autonomous vehicle tracking tasks.

Contribution

It proposes a novel particle-based score estimation approach using Fisher's identity, avoiding biased high-variance gradients from previous methods.

Findings

Learned models outperform existing techniques in accuracy.

Method provides more stable training for vehicle trajectory tracking.

Effective in real autonomous vehicle data scenarios.

Abstract

Multi-object state estimation is a fundamental problem for robotic applications where a robot must interact with other moving objects. Typically, other objects' relevant state features are not directly observable, and must instead be inferred from observations. Particle filtering can perform such inference given approximate transition and observation models. However, these models are often unknown a priori, yielding a difficult parameter estimation problem since observations jointly carry transition and observation noise. In this work, we consider learning maximum-likelihood parameters using particle methods. Recent methods addressing this problem typically differentiate through time in a particle filter, which requires workarounds to the non-differentiable resampling step, that yield biased or high variance gradient estimates. By contrast, we exploit Fisher's identity to obtain a particle-based approximation of the score function (the gradient of the log likelihood) that yields a low variance estimate while only requiring stepwise differentiation through the transition and observation models. We apply our method to real data collected from autonomous vehicles (AVs) and show that it learns better models than existing techniques and is more stable in training, yielding an effective smoother for tracking the trajectories of vehicles around an AV.