Spark

Robotic AI weighs risk and reward in hostile environments

If I go into that basin and never come out, at least you'll have the pictures!
A selfie by NASA's Mars Exploration Rover Opportunity, before it lost contact with Earth in June 2018. (NASA/Reuters)
Listen7:33

Since 2004, the Mars Opportunity rover travelled all around the Martian surface, investigating craters, getting stuck in difficult terrain and even helping NASA scientists discover evidence that there was once water on the Red Planet.

However, NASA lost contact with Opportunity in June 2018, and the space agency has now decided to say goodbye.

Opportunity had programming that gave it some autonomy to evaluate risk, overcome challenges and help NASA scientists make exciting discoveries.

Much of the AI we encounter evaluates risk in terms of making a decision, from giving you a loan to deciding whether to crest a difficult ridge on the surface of Mars.

But what about reward?  

For that, we need to leave outer space, and head down into the deep, dark sea.

Researchers watch footage taken by various AUVs (Autonomous Underwater Vehicles). (MIT )

"When we're throwing a robot into the ocean, there's a threat that it will collide with the environment," said Ben Aytona graduate student in MIT's Department of Aeronautics and Astronautics where he designs autonomous marine robots, in an interview with Spark host Nora Young. 

Several classes of autonomous underwater vehicles (AUVs) await deployment. (MIT)

But there's more to Ayton's job than just "throwing a robot into the ocean."

"What we're doing is trying to trade off this interplay between that threat of colliding with the environment and damaging the robot and potentially losing the robot, and the reward of what we see and what we gather." 

That means the robots can not only decide if a dangerous deep-sea situation is risky, but also whether it's worthwhile to continue an expedition altogether.

Comments

To encourage thoughtful and respectful conversations, first and last names will appear with each submission to CBC/Radio-Canada's online communities (except in children and youth-oriented communities). Pseudonyms will no longer be permitted.

By submitting a comment, you accept that CBC has the right to reproduce and publish that comment in whole or in part, in any manner CBC chooses. Please note that CBC does not endorse the opinions expressed in comments. Comments on this story are moderated according to our Submission Guidelines. Comments are welcome while open. We reserve the right to close comments at any time.