via Caltech
Engineers at Caltech, ETH Zurich, and Harvard are developing an artificial intelligence (AI) that will allow autonomous drones to use ocean currents to aid their navigation, rather than fighting their way through them.
“When we want robots to explore the deep ocean, especially in swarms, it’s almost impossible to control them with a joystick from 20,000 feet away at the surface. We also can’t feed them data about the local ocean currents they need to navigate because we can’t detect them from the surface. Instead, at a certain point we need ocean-borne drones to be able to make decisions about how to move for themselves,” says John O. Dabiri (MS ’03, PhD ’05), the Centennial Professor of Aeronautics and Mechanical Engineering and corresponding author of a paper about the research that was published by Nature Communications on December 8.
To tackle this challenge, researchers turned to reinforcement learning (RL) networks. Compared to conventional neural networks, reinforcement learning networks do not train on a static data set but rather train as fast as they can collect experience. This scheme allows them to exist on much smaller computers—for the purposes of this project, the team wrote software that can be installed and run on a Teensy—a 2.4-by-0.7-inch microcontroller that anyone can buy for less than $30 on Amazon and only uses about a half watt of power.
Using a computer simulation in which flow past an obstacle in water created several vortices moving in opposite directions, the team taught the AI to navigate in such a way that it took advantage of low-velocity regions in the wake of the vortices to coast to the target location with minimal power used. To aid its navigation, the simulated swimmer only had access to information about the water currents at its immediate location, yet it soon learned how to exploit the vortices to coast toward the desired target. In a physical robot, the AI would similarly only have access to information that could be gathered from an onboard gyroscope and accelerometer, which are both relatively small and low-cost sensors for a robotic platform.
This kind of navigation is analogous to the way eagles and hawks ride thermals in the air, extracting energy from air currents to maneuver to a desired location with the minimum energy expended. Surprisingly, the researchers discovered that their reinforcement learning algorithm could learn navigation strategies that are even more effective than those thought to be used by real fish in the ocean.
“We were initially just hoping the AI could compete with navigation strategies already found in real swimming animals, so we were surprised to see it learn even more effective methods by exploiting repeated trials on the computer,” says Dabiri.
Original Article: Engineers Teach AI to Navigate Ocean with Minimal Energy
More from: California Institute of Technology | ETH Zürich | Harvard University
The Latest Updates from Bing News & Google News
Go deeper with Bing News on:
Ocean-borne drones
- Giant military manta ray drone passes first ocean test
The five-year-old Manta Ray effort aims to harness the movement of the ocean to power a submersible drone that could stay out for far longer than current undersea drones. Such drones could help ...
- Ukraine’s balloon-borne bomber blitz: Designed to waste Russian shells and missiles
The latest Ukrainian deep-strike weapon isn’t a drone, a cruise missile or a ballistic missile ... Japan floated nearly 300 bomb-laden balloons across the Pacific Ocean in 1944 and ’45. The only ...
- Giant military manta ray drone passes first ocean test
An enormous underwater drone, built to roam the oceans for “very long periods of time” without refueling, has passed initial tests off California, DARPA announced on Wednesday. “Our successful, ...
- DARPA's autonomous 'Manta Ray' drone can glide through ocean depths undetected
Northrop Grumman Corporation has built its Manta Ray uncrewed underwater vehicle, which will operate long-duration missions and carry payloads into the ocean depths in partnership with DARPA.
- Best drones 2024: Expand your aerial horizons with the best drones
If you're searching for one of the best drones, at a range of price points to suit all budgets, then look no further than this comprehensive guide. Whether you're looking for a highly portable sub ...
Go deeper with Google Headlines on:
Ocean-borne drones
[google_news title=”” keyword=”ocean-borne drones” num_posts=”5″ blurb_length=”0″ show_thumb=”left”]
Go deeper with Bing News on:
Reinforcement learning
- Triomics taps LLMs to accelerate cancer care, raises $15M
In tests, Triomics' OncoLLM outperformed larger open-source and proprietary LLMs at patient-trial matching and rivaled qualified medical experts and GPT-4.
- [INTERVIEW] Merck will expand partnerships with Korean chip makers: executive VP
German chemical giant strives to repeat success in specialty gas bizMerck's Delivery Systems & Services (DS&S) and Specialty Gases (SG) businesses supply electronic materials and specialty gases ...
- OpenAI unveils specs for desired AI model behavior
The approach includes objectives, rules, and default behaviors that will guide OpenAI’s researchers and AI trainers who work on reinforcement learning from human feedback (RLHF). The company will also ...
- OpenAI unveils ‘Model Spec’: A framework for shaping responsible AI
The “Model Spec” draft document outlined a three-pronged approach to shaping AI behavior. “This document specifies OpenAI’s “desired model behavior” and how the company evaluates tradeoffs when ...
- DeepMind is experimenting with a nearly indestructible robot hand
A new robotic hand can withstand being smashed by pistons or walloped with a hammer. It was designed to survive the trial-and-error interactions required to train AI robots ...
Go deeper with Google Headlines on:
Reinforcement learning
[google_news title=”” keyword=”reinforcement learning” num_posts=”5″ blurb_length=”0″ show_thumb=”left”]