Grid-Based Object Tracking With Nonlinear Dynamic State and Shape Estimation

Abstract

Object tracking is crucial for planning safe maneuvers of mobile robots in dynamic environments, in particular for autonomous driving with surrounding traffic participants. Multistage processing of sensor measurement data is thereby required to obtain abstracted high-level objects, such as vehicles. This also includes sensor fusion, data association, and temporal filtering. Often, an early-stage object abstraction is performed, which, however, is critical, as it results in information loss regarding the subsequent processing steps. We present a new grid-based object tracking approach that, in contrast, is based on already fused measurement data. The input is thereby pre-processed, without abstracting objects, by the spatial grid cell discretization of a dynamic occupancy grid, which enables a generic multi-sensor detection of moving objects. On the basis of already associated occupied cells, presented in our previous work, this paper investigates the subsequent object state estimation. The object pose and shape estimation thereby benefit from the freespace information contained in the input grid, which is evaluated to determine the current visibility of extracted object parts. An integrated object classification concept further enhances the assumed object size. For a precise dynamic motion state estimation, radar Doppler velocity measurements are integrated into the input data and processed directly on the object-level. Our approach is evaluated with real sensor data in the context of autonomous driving in challenging urban scenarios.

Set-Based Prediction of Traffic Participants Considering Occlusions and Traffic Rules

“…We solely focus on motion prediction of other traffic participants [24]- [26], which is an integral part of motion planning [27]- [29] and risk assessment [24], [30]. The following related aspects are beyond the scope of this paper: extracting the information of surrounding traffic participants from sensor measurements [31]- [33], the uncertainty of these measurements [34]- [36], and implications on the prediction for connected vehicles [37], [38].…”

“…Therefore, we implemented our approach in C++ on a BMW 7 series test vehicle. The environment model provides the initial states of surrounding traffic participants based on [33] and the rectangular field of view without occlusions that extends 100 m in longitudinal and 60 m in lateral direction of the current pose of the ego vehicle. We use the planner of [15] to obtain trajectories for the ego vehicle that are collision-free against all predicted occupancies and bring the ego vehicle to a standstill; for the few cases the initial velocity is too high to come to a standstill within the planning horizon, we constrain the final state to comply with safe distances to predicted traffic participants.…”

A Grid-Based Framework for Collective Perception in Autonomous Vehicles

“…Vulnerable Road Users (VRUs), such as pedestrians and cyclists, are classified as other dynamics objects in this work. Nevertheless, the object classification can be extended to other categories as is shown in [ 41 ].…”

Performance Evaluation of the Highway Radar Occupancy Grid

“…This approach was first proposed by [ 10 ] for robotic applications. The procedure was extended into automotive applications by [ 15 , 16 ] for collision risk estimation and by [ 14 ] for grid-based object tracking. Nonetheless, previous evaluation methods are applicable only for specific test drive scenarios.…”

“…In the literature, there are metrics for object tracking [ 14 ] and for collision risk estimation [ 15 , 16 ]. In the object extraction steps of these methods, the occupancy grid must be carefully analysed and compared with accurate scene descriptions.…”