Papers
13
Total Citations
131
H-Index
7
About
Ransalu Senanayake is a robotics and machine learning researcher whose work spans probabilistic environmental mapping, autonomous navigation, and multi-robot systems. His most significant contributions center on advancing continuous occupancy mapping — moving beyond the limitations of traditional discretized grid maps to develop spatially and temporally coherent representations of dynamic environments. His foundational work on Spatio-Temporal Hilbert Maps and Bayesian Hilbert Map fusion has garnered substantial recognition, with key papers accumulating dozens of citations and establishing him as a leading voice in probabilistic robotics. Senanayake's research on dynamic Gaussian process occupancy maps and stochastic variational inference demonstrates a sophisticated integration of Bayesian methods with real-world robot navigation challenges, particularly in complex urban settings. His pedestrian flow modeling work further extends these ideas toward human-aware service robotics. More recently, he has broadened his scope to tackle multi-robot coordination through game-theoretic swarm control and policy learning via optimal transport in deep reinforcement learning. With roots in physical robot design — including early work on inchworm locomotion and subsea inspection manipulators — Senanayake's career reflects a rare breadth, bridging mechanical design and cutting-edge probabilistic learning to advance truly capable autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1Continuous Occupancy Map Fusion with Fast Bayesian Hilbert Maps26 citations · 2019
- 2Learning highly dynamic environments with stochastic variational inference23 citations · 2017
- 3Time-varying Pedestrian Flow Models for Service Robots17 citations · 2019
- 4
- 5Design and Development of a Planar Inchworm Robot11 citations · 2000
- 6
- 7Building Continuous Occupancy Maps With Moving Robots8 citations · 2018
- 8
- 9Online Domain Adaptation for Occupancy Mapping4 citations · 2020
- 10