CIMON: World’s first AI robot in Space

Reading Time: 4 minutes
4.8 (95%) 4 votes

When a character from science fiction comes to life, it shows how human beings have come so far as a civilization. Prof. Simon Wright, a flying human brain, is a character from the 1940s pulp magazine Captain Future which formed the basis of the free flying AI… Featured image is for representative purpose only and has been sourced from https://www.flickr.com/photos/dlr_de/28300532237

***

“You cannot create new science unless you realize where the old science leaves off and new science begins, and science fiction forces us to confront this.”

                                                                                                                            Dr. Michio Kaku

(American theoretical physicist, futurist and popularizer of science. He is a professor of theoretical physics in City College of New York and CUNY Graduate Center)

When a character from science fiction comes to life, it shows how human beings have come so far as a civilization. Prof. Simon Wright, a flying human brain, is a character from the 1940s pulp magazine Captain Future which formed the basis of the free flying AI Robot, CIMON.

CIMON, short for Crew Interactive Mobile Companion, is the latest addition to the International Space Station (ISS). CIMON is the first artificially intelligent robot to hover in zero G. It is designed and developed in a collaboration by three organizations – Airbus, DLR and IBM – to assist astronauts on space missions. While IBM has developed the AI core through its Watson System for information retrieval and natural language processing, German Aerospace Center (DLR) contracted Europe-based Airbus to manufacture the AI robot. Airbus has equipped this robot with automatic direction finding and object detection to freely float on the ISS without colliding with surrounding equipment or astronauts.

In deep space missions astronauts can face many challenges. Social interaction with the crew is of prime importance during these missions. This is where CIMON comes to the rescue with information and procedural data to solve concerns faced aboard the ISS. This gives astronauts the quick solution instead of waiting for directions from the team on Earth.

CIMON has a velocity of one meter per second smoothly hovers around the space station avoiding collision with surrounding objects and people.

How CIMON reached the ISS?

On 9th March 2018, CIMON underwent its first micro-gravity test for its autonomous navigation. Its 14 internal fans were tweaked constantly to maintain the robot in mid-air.

On 29th June 2018, CIMON was launched to the International Space Station (ISS) by SpaceX’s Falcon 9 rocket. The robot was a part of the several science experiments that were onboard SpaceX’s Dragon capsule. After completing a two-day trip CIMON arrived on the ISS on 2nd July 2018.

What is CIMON’s role?

The AI Robot will assist a European Space Agency astronaut, Alexander Gerst, during his stay on the ISS on three experiments. These experiments involve solving a Rubik’s Cube, crystal analysis and medical experiments. The ability to self-learn based on the environment and situation will be a great advantage in carrying out tasks. CIMON will constantly record these experiments and sync the data to the cloud for the ground team to analyze results.

If successful, CIMON will become the eyes and ears of the ground staff while constantly assisting astronauts on the ISS.

IBM’s IP finds its nemesis in CIMON

IBM is an active researcher of Natural Language Processing (NPL) that forms the basis of the AI installed in CIMON. In 2018, IBM published a patent US9892192 that discloses a question and answer system through which a natural language analysis is performed to select the most relevant set of questions, and answer them according the AI learning models.

Figure 3 of  US9892192 illustrates a simplified flow chart showing the logic for dynamically assigning question priority based on context information.

Another patent published by IBM in 2018 US9910848B2 describes a method for generating a question and answer pair using a natural language processing engine. Types of semantic are stored and a data processing system analyses the input to give the correct answer and according to the type of semantic a semantically correct answer is given by the system.

IBM’s 2017 patent publication US9767094 discloses a method to analyze a question/answer pair list from a user. The input from the user is processed by a natural language processing system which identifies the semantic type of the input. With the help of a data processing system, it performs a rule-based expansion of the input received from the user. The natural processing system implements a question/answer learning model to adapt to the inputs given by the user. Thus, this system constantly improves itself each time it interacts with the user.

Figure 4 of US9767094 illustrates a block diagram of a type-based semantic variant generation component in accordance with an illustrative embodiment.

CIMON with its numerous advantages in space travel will be an asset during future manned missions to the moon and Mars. The ability of CIMON to adapt to any situation and requirement of an astronaut makes it an indispensable tool for deep space missions. With upgrades being planned, CIMON will be leveraged for longer space missions as a social interactive guide for the ISS crew.

Featured image is for representative purpose only and has been sourced from https://www.flickr.com/photos/dlr_de/28300532237

Shravan Dua
Shravan Dua

An aeronautics enthusiast and space tech geek, Shravan loves indulging in books on astrophysics. When he is not busy honing his interest in science, playing the piano is what brings happiness to him.


2 Comments

  1. This Blog is great, and I never seen a Blog like this…..I really be grateful for you work and Thanks for providing us a great Platform for share information’s with others.

Post a Comment

Your email address will not be published. Required fields are marked *