A Random Finite Set Approach for Dynamic Occupancy Grid Maps with Real-Time Application

TL;DR

This paper introduces a novel random finite set-based filter for dynamic occupancy grid mapping, enabling real-time environment perception by fusing laser and radar data with efficient parallel particle filtering.

Contribution

It proposes the PHD/MIB filter, modeling environment dynamics as a stochastic system with multiple obstacles, and demonstrates a real-time fusion application with efficient implementation.

Findings

Parameters influence filter results as predicted theoretically.

Appropriate models yield consistent state estimates.

Real-time fusion of laser and radar data is achieved.

Abstract

Grid mapping is a well established approach for environment perception in robotic and automotive applications. Early work suggests estimating the occupancy state of each grid cell in a robot's environment using a Bayesian filter to recursively combine new measurements with the current posterior state estimate of each grid cell. This filter is often referred to as binary Bayes filter (BBF). A basic assumption of classical occupancy grid maps is a stationary environment. Recent publications describe bottom-up approaches using particles to represent the dynamic state of a grid cell and outline prediction-update recursions in a heuristic manner. This paper defines the state of multiple grid cells as a random finite set, which allows to model the environment as a stochastic, dynamic system with multiple obstacles, observed by a stochastic measurement system. It motivates an original filter called the probability hypothesis density / multi-instance Bernoulli (PHD/MIB) filter in a top-down manner. The paper presents a real-time application serving as a fusion layer for laser and radar sensor data and describes in detail a highly efficient parallel particle filter implementation. A quantitative evaluation shows that parameters of the stochastic process model affect the filter results as theoretically expected and that appropriate process and observation models provide consistent state estimation results.