CHANNELS

Aerospace

Automotive

Automated & Autonomous Vehicles

Battery & Electrification Technology

Defense

Drone TV

Electronics

Manufacturing & Prototyping

Materials

Medical

Motion Control

RF & Microwave Technology

Robotics & Automation

Sensors & IoT

Test & Measurement

Aerospace
Automotive
Automated & Autonomous Vehicles
Battery & Electrification Technology
Defense
Drone TV
Electronics
Manufacturing & Prototyping
Materials
Medical
Motion Control
RF & Microwave Technology
Robotics & Automation
Sensors & IoT
Test & Measurement
Aerospace & Defense
Search Results

LINCing on the Charles River for AI-Enabled Safety Control Tests

The Learning Introspective Control (LINC) program aims to develop machine learning-based introspection technologies that enable physical systems — with specific interest in ground vehicles, ships, drone swarms, and robotic systems — to respond to events not predicted at design time. Watch this video to see LINC tested onboard a vessel on the Charles River in Boston.

Topics:
AR/AI Artificial Intelligence Inspection Equipment Measuring Instruments Monitoring Test & Measurement Testing Procedures

More From SAE Media Group

    Transcript

    00:00:00 the link program learning introspective control has the aspiration of being able to give mechanical systems the ability to suffer and then recover from damage or disruptions in an automated way without pre-learning an example is you're driving a truck down a road you have a blowout it begins to fish tail if you're a very skilled driver you can get it back under control and get it safely

    00:00:26 to the side of the road if you're driving a link adapted truck it'll figure out oh things are wrong adjust the steering and the braking where it feels like the truck has not been disturbed but it's able to drive safely to the side of the road doing the same thing with ships and airplanes is part of the link program we know that vehicles change during the course of a

    00:00:46 mission and we saw one today that we didn't expect and that was the failure of a of a thruster and the thruster just stopped working because it happened to wear out at that particular moment in time and part of the things we're exploring in this program is how we can detect that and how we can compensate for that next we performed a scenario with two of these vessels simulating a

    00:01:05 subscale underway replenishment mission in the scenario there were multiple disturbances acting on the vessels there were thruster failures there were simulated interaction forces between the two vessels and there was a deployable drag device which was a bucket when we tested the baseline controller the system was unable to recover from all of those simultaneous disturbances however

    00:01:29 when we implemented the link controller the system was able to recover very quickly and complete the mission successfully there are two things that excite me the most the fact that we can have systems that can recover and protect and save people that essentially give expert skills to all operators and that can be creative and innovative in the way it finds solutions the second

    00:01:50 one is it allows us to extend the operating life of systems which means we don't have to manufacture as many things we don't have as many things going to waste and it can really drop the long-term carbon footprint of our systems we'll have longer lived safer more reliable systems and that's what's excites me