Divergence (linguistics)
Related papers: 20
About
Divergence, in the context of robotics and AI, refers to a family of mathematical and perceptual measures that quantify how much two distributions, signals, or flow fields differ from one another. It appears in two distinct but related forms. The first is statistical divergence, most notably Kullback-Leibler (KL) divergence, which measures the difference between probability distributions and is widely used in reinforcement learning for constraining policy updates, in multi-sensor fusion for fault detection, and in active SLAM for guiding exploration toward maximally informative robot motions. The second is optical flow divergence, a visual cue derived from expanding or contracting patterns in an image stream that allows robots and aerial vehicles to estimate time-to-contact, surface orientation, and proximity to obstacles without explicit depth sensing. Both forms matter because they provide principled, computationally efficient signals: statistical divergence keeps learned policies stable and well-calibrated, while optical flow divergence enables lightweight, biologically inspired navigation in resource-constrained platforms such as micro air vehicles and mobile robots operating in real time.
Top Researchers
Top Institutes
Top Cited Papers
Consistent cooperative localization
Alexander Bähr, M. R. Walter, John J. Leonard
Citations: 186 • 2009
Real-time obstacle avoidance using central flow divergence and peripheral flow
David Coombs, Martin Herman, Tsai Hong Hong, Marilyn Nashman
Citations: 170 • 1995
Surface orientation and time to contact from image divergence and deformation
Roberto Cipolla, Andrew Blake
Citations: 165 • 1992
Ensemble-CIO: Full-body dynamic motion planning that transfers to physical humanoids
Igor Mordatch, Kendall Lowrey, Emanuel Todorov
Citations: 145 • 2015
Concurrent stiffening and softening in hydrogels under dehydration
Shuai Xu, Zidi Zhou, Zishun Liu, Pradeep Sharma
Citations: 134 • 2023
Multi-sensor fusion approach with fault detection and exclusion based on the Kullback–Leibler Divergence: Application on collaborative multi-robot system
Joelle Al Hage, Maan E. El Najjar, Denis Pomorski
Citations: 121 • 2017
Constrained Policy Optimization
Joshua Achiam, David Held, Aviv Tamar, Pieter Abbeel
Citations: 112 • 2017
Robust and efficient robotic mapping
Seth Teller, John J. Leonard, Edwin Olson
Citations: 109 • 2008
Real-time obstacle avoidance using central flow divergence, and peripheral flow
David Coombs, Martin Herman, Tsai-Hong Hong, Marilyn Nashman
Citations: 103 • 1998
Active SLAM and Exploration with Particle Filters Using Kullback-Leibler Divergence
Luca Carlone, Jingjing Du, Miguel Kaouk Ng, Basilio Bona, Marina Indri
Citations: 98 • 2013
Real-time single-workstation obstacle avoidance using only wide-field flow divergence
Ted Camus, David Coombs, Martin Herman, Tsai-Hong Hong
Citations: 74 • 1996
Variational Inference for Policy Search in changing situations
Gerhard Neumann
Citations: 70 • 2011
Monocular distance estimation with optical flow maneuvers and efference copies: a stability-based strategy
Guido de Croon
Citations: 58 • 2016
Autonomous convergence and divergence of the self-powered soft liquid metal vehicles
Jie Zhang, Youyou Yao, Jing Liu
Citations: 56 • 2015
Online walking pattern generation for push recovery and minimum delay to commanded change of direction and speed
Junichi Urata, Koichi Nshiwaki, Yuto Nakanishi, Kei Okada, Satoshi Kagami, Masayuki Inaba
Citations: 49 • 2012
Model-based imitation learning by probabilistic trajectory matching
Péter Englert, Alexandros Paraschos, Jan Peters, Marc Peter Deisenroth
Citations: 49 • 2013
Neutral networks in an evolutionary robotics search space
Tom Smith, Phil Husbands, Michael O’Shea
Citations: 49 • 2002
Robust Contact Generation for Robot Simulation with Unstructured Meshes
Kris Hauser
Citations: 48 • 2016
Adaptive Step-Size for Policy Gradient Methods
Matteo Pirotta, Marcello Restelli, Luca Bascetta
Citations: 48 • 2013
Factored four way conditional restricted Boltzmann machines for activity recognition
Decebal Constantin Mocanu, Haitham Bou Ammar, Dietwig Lowet, Kurt Driessens, Antonio Liotta, Gerhard Weiß, Karl Tuyls
Citations: 47 • 2015