Abstract:Grid-centric perception is a crucial field for mobile robot perception and navigation. Nonetheless, grid-centric perception is less prevalent than object-centric perception as autonomous vehicles need to accurately perceive highly dynamic, large-scale traffic scenarios and the complexity and computational costs of grid-centric perception are high. In recent years, the rapid development of deep learning techniques and hardware provides fresh insights into the evolution of grid-centric perception. The fundamental difference between grid-centric and object-centric pipeline lies in that grid-centric perception follows a geometry-first paradigm which is more robust to the open-world driving scenarios with endless long-tailed semantically-unknown obstacles. Recent researches demonstrate the great advantages of grid-centric perception, such as comprehensive fine-grained environmental representation, greater robustness to occlusion and irregular shaped objects, better ground estimation, and safer planning policies. There is also a growing trend that the capacity of occupancy networks are greatly expanded to 4D scene perception and prediction and latest techniques are highly related to new research topics such as 4D occupancy forecasting, generative AI and world models in the field of autonomous driving. Given the lack of current surveys for this rapidly expanding field, we present a hierarchically-structured review of grid-centric perception for autonomous vehicles. We organize previous and current knowledge of occupancy grid techniques along the main vein from 2D BEV grids to 3D occupancy to 4D occupancy forecasting. We additionally summarize label-efficient occupancy learning and the role of grid-centric perception in driving systems. Lastly, we present a summary of the current research trend and provide future outlooks.

Performance Evaluation of the Highway Radar Occupancy Grid

Dynamic Occupancy Grids for Object Detection: A Radar-Centric Approach

A Quantitative Evaluation For Radar Grid Map Construction

A Fusion Method Aiming at Environmental Perception of Autonomous Vehicle Based on Visual Scheme

Precise Occupancy Grid Map Generation Using Only FMCW Radar

Deep Radar Inverse Sensor Models for Dynamic Occupancy Grid Maps

Moving Objects Detection in Evidential Occupancy Grids Using Laser Radar

Scalable Radar-based Roadside Perception: Self-localization and Occupancy Heat Map for Traffic Analysis

Vehicle Detection on Occupancy Grid Maps: Comparison of Five Detectors Regarding Real-Time Performance

A New Grid Map Construction Method for Autonomous Vehicles

Road-Map Aided GM-PHD Filter for Multivehicle Tracking With Automotive Radar

Collaborative Grid Mapping for Moving Object Tracking Evaluation

Radar Detection Rate Comparison through a Mobile Robot Platform at the ZalaZONE Proving Ground

Road Boundaries Detection based on Modified Occupancy Grid Map Using Millimeter-wave Radar

Grid-Centric Traffic Scenario Perception for Autonomous Driving: A Comprehensive Review

Occupancy Grid Mapping Based on DSmT for Dynamic Environment Perception

RadarOcc: Robust 3D Occupancy Prediction with 4D Imaging Radar

Raw High-Definition Radar for Multi-Task Learning

Automated Automotive Radar Calibration With Intelligent Vehicles

The Radar Ghost Dataset -- An Evaluation of Ghost Objects in Automotive Radar Data