Skip to content

IEEE-ESSTHS/YP-Challenge

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Phoebe RL Agent

Phoebe is a reinforcement learning agent designed to optimize and help astronauts and NASA ground control to perform various decision-making tasks on the ARTEMIS Lunar Mission.

Phoebe is built with Python, Flask, and Vue.js, it is built to be compatible with NASAs open mission control platform, see more here

Phoebe performs Multiple Tasks

Installation

Use the package manager pip to install packages from requirements.txt.

pip install -r requirements.txt

to run this api

python app.py

Usage

This section is a flask API for the Phoebe computation and responses Phoebe takes an image as in input, which is the mars surface seen by the VIPER rover This image is segmented into multiple tiles and converted to a Numpy array taking into consideration the terrain changes Phoebe agent then tries different policies on each segment to deduct the optimal path between each tile on that image corresponding to the moon surface zones.

Phoebe also has a simple task rearrangement based on priority and given conditions like ( sun position, atmosphere state, astronaut physiological shape, priority, and unexpected conditions ) to change the order and priority of the different tasks required to be done by the astronauts.

API endpoints

# '/'
Returns a JSON response of basic info about the Phoebe project

# '/path/'
Returns the optimal path for each segment of the last inputted image by the VIPER rover

# '/tasks/'
Updates the last tasks updates depending on the various inputs and conditions given 
and returns newly updated task lists to arrange them in the interface

Contributing

Pull requests are welcome. For significant changes, please open an issue first to discuss what you would like to change.

Please make sure to update tests as appropriate.

Reinforcement Learning Approach for the Phoebe Agent

Simplified Image for how Reinforcement Learning agents work

RL

Phoebe was based on the Monte Carlo algorithm

Algorithm

Policy Iterations help us pick the optimal policy for each tile Policy

Frontend

Dashboard

Tasks

External Sources

LUNANET: Openmct: Open MCT (Open Mission Control Technologies) is a next-generation mission control framework for visualization of data on desktop and mobile devices

[1]https://www.nasa.gov/feature/goddard/2021/lunanet-empowering-artemis-with-communications-and-navigation-interoperability [2]https://independentaustralia.net/business/business-display/lunanet-developing-internet-for-the-moon,16716 [3]https://github.com/nasa/openmct

Volatiles Investigating Polar Exploration Rover (VIPER) [4]https://www.nasa.gov/viper/lunar-operations#route [5]https://www.nasa.gov/feature/ames/nasas-viper-prototype-motors-through-moon-like-obstacle-course

Base Camp Management [6]https://www.smithsonianmag.com/science-nature/four-things-weve-learned-about-nasas-planned-base-camp-on-the-moon-180980589/ [7]https://www.smithsonianmag.com/air-space-magazine/tools-of-the-astronaut-trade-15273242/ [8]https://blogs.nasa.gov/artemis/2020/10/28/lunar-living-nasas-artemis-base-camp-concept/

License

MIT