Author - Steve Jones

1
Crowd sourcing deep space communication
2
How Deep Learning could revolutionise Earth Observation

Crowd sourcing deep space communication

blog photo

Deep space communication poses a range of significant challenges. To overcome them, NASA is taking more flexible approaches to problem solving than ever before. We look at how their collaboration with TopCoder is evolving the uses of crowd-sourced code, and consider what lies ahead

Email is so ubiquitous and fast that we may be inclined to take it for granted, but what if you were trying to email home from another planet? In this context – or the ‘Matt Damon scenario’ as I now call it – a number of issues may cause you problems, including distance, planet rotation and transmission power limits. These obstacles are not just hypothetical; astronauts on the International Space Station often suffer email issues, particularly when attempting to send large attachments. To achieve effective deep space communication, space agencies therefore need to build what are known as disruption-tolerant networks.

TopCoder has hosted technical competitions since 2001 and they’re understandably considered to be at the premium online technical community. Members select contests and view individual contest requirements. Specialists within the community then compete in a series of competitions that comprise the whole project.

In April 2014, NASA launched a project with TopCoder to improve computer network architecture for deep space, and it completed in June 2015. In total, there were 12 challenges they were worked on by 146 contributors from 42 countries.

Bundle of fun

So what exactly were the coders working on? The goal was ultimately to provide a shared framework for algorithm and application development in disruption-tolerant networks.

NASA had previously published a couple of ‘requests for comment’ on an experimental protocol (commonly known as the Bundle Protocol) they were running on disrupted networks, which defines a series of contiguous data blocks as a bundle. Each of these bundles provides sufficient semantic information for the application to make progress where an individual block might not.

Overcoming disruption issues would also help improve security as authentication and privacy are often critical to data sent via disruption-tolerant networks. These security guarantees are difficult to establish in a network without persistent connectivity: the disruptions to the network can hinder complicated cryptographic protocols and key exchange, and each device must identify other intermittently visible devices.

Successful solutions

The solution the coders developed includes both client- and server-sides of the communication. Importantly, the support code doesn’t interfere with ground users using the same exchange server. The code supports unpredictable suspension of communication for up to four hours, unpredictable loss of data, and round-trip times on the order of .6 s – 1 s. In short, the project outputs ably met NASA’s requirements.

This isn’t the first collaboration between NASA and TopCoder. In fact, the NASA National Tournament Lab was actually established in 2011 as a result of a 2009 TopCoder challenge. Clearly then, NASA sees this crowd sourcing approach as broadly beneficial. However, while this challenge won’t be the last, in the realm of deep space communications, NASA seems intent on prioritising lasers.

According to Kevin Carmack, NASA’s Laser Communications Relay Demonstration Project Manager, lasers represent the “optical communication of the future”. This new technology is already proving to work as effectively as many in the sector believed it would. As such, the gains in bandwidth and the consequent speed of large data transference, means disruption-tolerant networks may soon be rendered a thing of the past. Not only do lasers require less transmission power, the low divergence of laser beams makes them a secure option for long range communications.

New miniature OCSD satellite launches

To help test the potential of laser communication on small scale satellites, NASA and The Aerospace Corporation of El Segundo, California have just launched the Optical Communications and Sensor Demonstration (OCSD) CubeSat. The innovation here is that the laser is hard-mounted to the spacecraft body, and the orientation of the CubeSat controls the direction of the beam, allowing design of the most compact system ever.

Along with other government agencies, academia and commercial companies, NASA can use the results of the test to incorporate laser communication technology into future space missions, as Steve Jurczyk, associate administrator for NASA’s Space Technology Mission Directorate states: “Technology demonstration missions like OCSD are driving exploration. By improving the communication capability of small spacecraft to support data-intensive science missions, OCSD will advance the potential to become a more viable option for mission planners.”

With the rapid speed of technology convergence, it therefore seems that NASA has started to understand how to allow open innovation to flourish more readily within Administration projects.

How Deep Learning could revolutionise Earth Observation

earth-and-satellites

Earth Observation has been a little overlooked this year but integrating Deep Learning technologies could power rapid developments across the entire field. Here, we look at two interesting ways this sub-domain of artificial intelligence could aid industry and international agencies alike.

It’s hard to argue with the enthusiasm for looking upwards and outwards that has swept the globe in recent months; we have witnessed some truly historic scientific landmarks, including the discovery of water on Mars and the emphatic flyby of Pluto (NASA is releasing new images of the dwarf planet’s blue sky as I type). Of course, it’s worth remembering that those achievements came about from innovative technologies converging over the past decade, and similar forward leaps look set to revolutionise Earth Observation in the near future.

Earth Observation and image recognition

When looking down from above, one longstanding challenge with managing satellite data concerns hyperspectral data classification. For a moment, however, let’s look at a broader level and consider image recognition technology more generally.

Historically, programming computers to classify images effectively has been difficult. Google Image Search demonstrates this point well; this search engine does not currently recognise the image itself, but finds matches by analysing metatags and page content, the origins of which have at some point been manually user-generated. Looking to the future, however, these search functions will be able to match images by actually understanding what an object looks like and matching that with others it understands to be similar. In fact, Baidu, the Chinese search giant, and others, have already arrived at this point. This area of work, known as Deep Learning, is most commonly based on creating convolutional neural networks (if you’re an outsider to the field, they might sound incredibly complicated, but don’t worry, to an insider they seem even more complex). In essence, though, they are systems by which computer agents learn to recognise image segments at a minute scale as a stepping stone to understanding features and then finally objects and scenes, without additional inputs beyond raw data.

Deep Learning and hyperspectral data classification

Returning to an Earth Observation context, hyperspectral data are essentially images that reveal added context about the materials pictured. Classification systems for these images are ineffective at present and huge mountains of data are being constantly harvested. Fortunately, we seem to be right at the upswing of exponential improvements here. And, such is the nature of Deep Learning, once engineers reach a certain level of success with algorithm development for the agent itself, raw data is all that is required for further improvement. Usefully, raw data is not something satellites struggle to collect, they do so in abundance, which leads neatly on to Deep Learning applications and data storage.

Can Deep Learning help with data storage?

So what about data storage? Well here’s where things get a little more theoretical, but also more fascinating with regards to artificial intelligence. Some researchers, have described a new type of deep neural network – a Perpetual Learning Machine – that uses statistical recall biases to mimic human memory to a limited extent, creating a ‘use it or lose it’ stimulus driven memory process. The suggestion is that researchers could build algorithms that make decisions about what to use their memory for, essentially ‘forgetting’ information they have learnt to deem unimportant through experimentation on huge datasets. The implication is that the huge challenges in data storage that currently exist could be overcome not by us, but by computerised agents themselves.

Going deeper

This only scrapes the surface of some very complex fields of course, but it does go some way to highlighting just how exciting near-future technology convergence could be for Earth Observation applications. If hyperspectral data classification or storage are important to your project or business, but you don’t work in these research fields directly, I’d suggest seeking out the best minds in Deep Learning right now, as there are still so few who really grasp the subject area at both broad and detailed scales.

Keep up to date

Twitter is a great place to keep up to date with Earth Observation developments as they happen. Follow us here:

Tweets by @spacecongress

To save you time, we’ve also curated a list of other top Earth Observation Twitter accounts to follow:

• Aberystwyth University: @AU_EarthObs
• African Physicists: @AfricanPhysics
• Canadian Space Agency: @CSA_ASC
• Disasters Charter: @DisastersChart
• Earth Observing Lab: @ncareol
• ESA EarthObservation: @ESA_EO
• ESA Italy: @ESA_Italia
• ESA Science: @esascience
• European Space Agency: @ESA
• Group on Earth Observations: @GEOSEC2025
• India Space: @India_inSpace
• Japan Aerospace Exploration Agency: @JAXA_en
• NASA Astronauts: @NASA_Astronauts
• NASA EO: @NASA_EO
• NASA ESTO: @NASAESTO
• NASA Goddard: @NASAGoddard
• NASA Landsat Program: @NASA_Landsat

NASA Soil Moisture Active Passive mission @NASASMAP

 

• NASA: @NASA
• National Weather Service: @NWS
• NOAA: @NOAA
• Sensing Our Planet: @sensingrplanet
• Space Generation Advisory Council: @SGAC
• The World Meteorological Organization: @WMOnews
• UK Space Agency: @spacegovuk
• World Resources Institute: @worldresources