Project Proposal

Introduction

This document serves as the project proposal for Computational Biology CS502 and Virtual and Augmented Reality CS491 courses at UIC. Our group is composed by Matteo Foglio and Andrea Milanta, graduate students enrolled in the joint master degree program
between UIC and Politecnico di Milano,

The purpose of the project is to evaluate two biological models of the the foraging process of a particular species of monkeys and provide an effective visualization tool to validate the results. Our models will be tested on data acquired from spider monkeys on Barro Colorado Island.

Background

Barro Colorado Island is an island in the middle of the Panama Canal devoted to natural research since 1923. In the 1960s a small population of young spider monkeys (Ateles Geoffroyi) was introduced in the island and has been monitored since [1]. The spider monkey is characterized by a highly arboreal lifestyle, meaning that they spend most of their time on top of the trees and rarely descend to the ground [2]. The spider monkey feeds on the fruit of some trees and move along the canopy in order to reach more food.

Moreover, in recent years thanks to the use of drone technology and a new data acquisition framework (SENSEI), the forest canopy has been precisely mapped and a lot of data on monkey movement has been acquired using GPS technology.

Visualization projects are also underway to help visualize and better understand the acquired data. One of the most prominent is SENSEI-Panama. Conducted at the University Of Illinois At Chicago, it aims at visualizing the path of the monkeys on the canopy in their state-of-the-art CAVE2 virtual reality environment [3][4].

Ateles Geoffroyi
Ateles Geoffroyi

 

 

 

 

 

Proposal

The aim of our project is to understand the behavior of the monkeys in the foraging process. More in detail we want to understand if the movement of the monkeys toward a fruit tree is driven by the sighting of it or by some previous knowledge of its position.

We plan to divide the project into two main branches. The first one will deal with data analysis and creation of models to be validated against the acquired data. The second one will focus on porting the SENSEI Panama project into Unity and develop a more effective display of the data and the result. This can be very important to provide a concrete proof of the correctness of our results.

Data Analysis and Machine Learning

We propose two models to be validated:

  1. The monkeys randomly roam around the canopy until they see a tree with fruit. Once they have seen one, they go towards it along the shortest possible path.
  2. The monkeys already know the position of the fruit tree and navigate directly to them without necessarily see them.

We will begin by analyzing the dataset of both monkey movements and the canopy. From that we will create two dataset of artificial monkey movements that fit the two models. We will then train one classifier for each dataset and check the real data against them.

Visualization

We plan to implement a virtual reality application to visualize the data acquired and visually verify the results of the model estimations.

The first step will be to port the current SENSEI Panama project, developed in OmegaLib at the UIC EVL, into Unity. The current implementation allows the user to have a bird's-eye view of the island where all the trees with fruit are highlighted. Using a simple menu it is possible to display movement information of one or multiple monkeys on the island. After the porting we plan to customize the monkey movement representation highlighting the factors that mostly influence the behavioral models.

We decide to port the project into Unity to take advantage of its advanced tools and provide a ground base that can be more easily expanded in the future. Moreover we plan to visualize the data using HTC Vive, against the original Cave2, which can be helpful to share the application to a broader audience.

Although not one of our main goal, we aim at producing a visual application that can be easily modified to display data coming from different researches.

References

[1] Milton K., Hopkins M.E. (2006) Growth of a Reintroduced Spider Monkey (Ateles geoffroyi) Population on Barro Colorado Island, Panama. In: Estrada A., Garber P.A., Pavelka M.S.M., Luecke L. (eds) New Perspectives in the Study of Mesoamerican Primates. Developments in Primatology: Progress and Prospects. Springer, Boston, MA

[2] Campbell, C.J., Aureli, F., Chapman, C.A. et al. Int J Primatol (2005) 26: 1039. https://doi.org/10.1007/s10764-005-6457-1

[3] Hwang J., Aurisano J, San-Juan O., Bhagat I. SENSEI-Panama Project, Electronic Visualization Laboratory (EVL) at the University of Illinois at Chicago (UIC) https://www.evl.uic.edu/entry.php?id=2228

[4] CAVE2: Next-Gneration Virtual-Reality and Visualization Hybrid Environment for Immersive Simulation and Information Analysis https://www.evl.uic.edu/entry.php?id=2016