Learning Occupancy Grid Maps with Forward Sensor Models

By Sebastian Thrun. Auton. Robots, 2003

561 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Networks and Communications, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/7fbac3c51867e514b287052e2519578bfa96cb45

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Networks and Communications

In Computer Vision and Pattern Recognition

Citations in the corpus

Exploiting the Power of Stereo Confidences (2013, methodology): “By using the pout parameter this forward sensor model [23] allows to consider that a disparity value might be an outlier.”
Robust Probabilistic Occupancy Grid Estimation from Positive and Negative Distance Fields (2012, methodology): “Probabilistic occupancy grids have arguably been the dominant paradigm for map building in robotics [25, 4, 32] partly becasue they are a world-based representation that enables the fusion of measurements from mobile sensors, even in the presence of large uncertainties in motion estimation.”
Algorithms for collision-free navigation of mobile robots in complex cluttered environments: a survey (2014, background): “Mapping is becoming very popular in real-world applications, where exploration of unknown environments is required (see, e.g., [71, 115, 272]).”
VoxNet: A 3D Convolutional Neural Network for real-time object recognition (2015, methodology): “Using the formulation from [31], we update each voxel traversed by the beam as l t ijk = l t−1 ijk + z t l occ + (1 − z t )l free (1) where l occ and l free are the log odds of the cell being occupied or free given that the measurement hit or missed the cell, respectively.”
MP3: A Unified Model to Map, Perceive, Predict and Plan (2021, methodology): “We follow the philosophy of [13,20,45,48,51] in predicting scene-level representations, but propose an improved occupancy flow parameterization that can model multi-modal behavior and provides a consistent temporal motion field.”
Segmentation of 3D lidar data in non-flat urban environments using a local convexity criterion (2009, background): “Most frequently, data is projected to an assumed or estimated ground plane, often combined with an occupancy grid map [ 5 ].”
Tracking Multiple Occluding People by Localizing on Multiple Scene Planes (2009, background): “In spite of the current body of knowledge, we believe monocular methods have limited ability to handle occlusions involving several objects, generally two or three, because the single viewpoint is intrinsically unable to observe the hidden areas.”
Multiple‐Robot Simultaneous Localization and Mapping: A Review (2016, methodology): “Polygon-based SLAM is used for 3D mapping (Hahnel, Burgard, & Thrun, 2003b).”