Research

Non-linear Methods for Modeling Uncertainty in Spatiotemporal Phenomena with Applications to Robotic Mapping (2015 - Present) 

 Occupancy mapping                                                                                     

  with A/ Prof. Fabio Ramos and Dr. Simon O'Callaghan

Temporal variations of spatial processes exhibit highly nonlinear patterns and modeling them is vital in many disciplines. For instance, robots operating in dynamic environments demand richer information for safer and robust path planning. In order to model these spatiotemporal phenomena, I develop and utilize theory in reproducing kernel Hilbert space (RKHS) and deep learning. I am mainly interested in modeling and propagating the uncertainty of dynamic environments and therefore I frequently use Bayesian modeling techniques.


Bayesian Hilbert maps

Faster Gaussian process mapping























 Keywords: Scalable Bayesian inference, RKHS, sparse Gaussian processes

Software/Tools/Methods: Python and C++



R. Senanayake and F. Ramos“Bayesian Hilbert Maps for Dynamic Continuous Occupancy Mapping, AAAI, 2018

R. Senanayake, T. Ganegedara, and F. RamosDeep Occupancy Maps: a continuous occupancy mapping technique for dynamic environments, NIPS-MLITS workshop, 2017

R. Senanayake, S. O'Callaghan, and F. RamosBayesian Hilbert Maps for Continuous Occupancy Mapping in Dynamic Environments, ICML-MLAV workshop, 2017

R. Senanayake and F. RamosContinuous Occupancy Mapping with Moving Robots, CoRL, 2017

R. Senanayake, S. O'Callaghan, and F. RamosLearning Highly Dynamic Environments with Stochastic Variational Inference,” ICRA, 2017

R. Senanayake, L. Ott, S. O'Callaghan, and F. Ramos
“Spatio-Temporal Hilbert Maps for Continuous Occupancy Representation in Dynamic Environments,
NIPS, 2016

R. Senanayake, S. O'Callaghan, and F. RamosMapping Occupancy of Dynamic Environments using Big Data Gaussian Process Classification, NIPS-MLITS workshop, 2016

R. Senanayake, S. O'Callaghan, and F. Ramos
“Predicting Spatio–Temporal Propagation of Seasonal Influenza Using Variational Gaussian Process Regression,” AAAI, 2016

Intel Lab


 Directional Mapping                                                                                     


Human Interaction in Targeted-Tracking with Pointing Devices (2012 - 2013) 

              



 Keywords: Human Factors


Considering open-loop and visual feedback control of hand/arm movements in combined targeting tracking tasks of the human-computer interface, a statistical model was developed. The model is a set of four equations derived based on statistical analysis of data collected from human subject based experiments. The model is important in:
  • Design and evaluation of input devices
  • Design and evaluation of graphical user interfaces (GUI)
  • Design and evaluation of virtual environments such as computer games
Additionally, the experimental results can be exploited to mimic hand movement dynamics in robotic manipulation as well. We developed a predictive model for targeted-tracking tasks. The model was experimentally verified for different input devices.





 



Publications:

R. Senanayake, and R.S. Goonetilleke,Pointing device performance in steering tasks, Perceptual and motor skills, 2016.  doi:10.1177/0031512516649717 [link]

R. Senanayake, R.S. Goonetilleke and E.R. Hoffmann, “Targeted-tracking with pointing devices”, IEEE Transactions on Human-Machine Systems, 2015 [link]

R. Senanayake, E.R. Hoffmann and R.S. Goonetilleke, “A model for combined targeting tracking tasks in computer applications”, Experimental Brain Research, 2014 [link]

R. Senanayake, and R.S. Goonetilleke, “Superiority of Freehand Pointing”, in Proceedings of the 57th Annual Meeting of the Human Factors and Ergonomics Society,  San Diego, USA, 2013 [link]

R. Senanayake, and R.S. Goonetilleke, “Setting that Mouse for Tracking Tasks”, in Proceedings of  the 15th International Conference on Human-Computer Interaction, Las Vegas, USA, 2013 [link]


Software/Tools/Methods:

The software used in the experiment was developed using C++ (with OpenCV)
Analysis software: Matlab, SPSS, Minitab
Statistical methods: Stepwise regression with bidirectional elimination, repeated measures ANOVA with Greenhouse-Geisser correction, principal component analysis, non-parametric post-hoc tests, etc.  


Other works: 
  • Developing eye-controlled computer games with Tobii REX (HTML5/JavaScript)

                                                                                                                                                                                                                                    More




Vision Based Hand Gesture Recognition for Appliance Control  (2011)                                                                    



We developed a method for appliance control based on hand gestures. The challenges were to make the system less prone to cluttered & dynamic background and not requiring the user to wear long-sleeved garments or wrist/finger bands. Our contribution was for the development of Smart Homes.

The method involved delimitating hand from rest of the background, extracting features that are invariant to translation/rotation and classification with neural networks.


 

Publications: 
R. Senanayake, and S.P. Kumarawadu, “A Robust Vision-based Hand Gesture Recognition System for Appliance Control in Smart Homes”, presented at the IEEE International Conference on Signal Processing, Communications and Computing, Hong Kong, 2012

Software/Hardware:

C/C++ (with OpenCV), Matlab, PIC controller (with USB HID communication)

 More                      

Automatic Detection of Malarial Parasites in Human Blood Smear (2015)   

      Student: Ahalya Ravendran

Publications:
 A. Ravendran, R.T. de Silva, and R. Senanayake, “Moment Invariant Features for Automatic Identi.cation of Critical Malarial Parasites”, presented at the 10th Int.  Conf.  on Industrial and Info.  Systems, Peradeniya, 2015 [link]

Computer-Controlled Digital Microscope (2011-2012)                                                                                           


 A low-cost semi-automated digital microscope was developed. Focusing and illumination control is software controlled. We incorporated many existing image processing algorithms for photomicrograph enhancement.

Publications:
A.D.N. Silva, M.N. Wijesundara, and R. Senanayake, “Computer Controlled Digital Microscope with Photomicrograph Enhancement”, presented at the IEEE International Conference on Information and Communication Technology, Indonesia, 2013