DARPA Robots Learn to Grasp Objects and Carry Loads Over Rough Terrain

Researchers with DARPA’s Machine Common Sense (MCS) program demonstrated a series of improvements to robotic system performance over multiple experiments. Just as infants must learn from experience, MCS seeks to construct computational models that mimic the core domains of child cognition for objects (intuitive physics), agents (intentional actors), and places (spatial navigation). Using only simulated…
DARPA Robots Learn to Grasp Objects and Carry Loads Over Rough Terrain


Researchers with DARPA’s Machine Common Sense (MCS) program demonstrated a series of improvements to robotic system performance over multiple experiments. Just as infants must learn from experience, MCS seeks to construct computational models that mimic the core domains of child cognition for objects (intuitive physics), agents (intentional actors), and places (spatial navigation).

Using only simulated training, recent MCS experiments demonstrated advancements in systems’ abilities – ranging from understanding how to grasp objects and adapting to obstacles, to changing speed/gait for various goals.

Nextbigfuture believes that work on humanoid robots will result in robots that can walk and pick up and place objects at human levels within a couple of years. This will result in a massive change in world economics. I describe the impact in this video.

The humanoid bots will work with the millions of existing industrial robots and the millions of existing warehouse robots.

Oregon State researchers demonstrated the ability for a bipedal robot to learn how to carry dynamic loads with only proprioceptive feedback. The robot, known as Cassie, learned commonsense behaviors in a simulated-to-real learning environment. Cassie adapted its gait to account for changes in load dynamics, such as sloshing liquids or balancing weights. After training in simulation, Cassie was able to walk on a treadmill for several minutes with four different types of dynamic loads. In contrast, before the learned commonsense training, Cassie fell immediately.

University of Utah researchers as part of the Oregon State University MCS team developed an active, grasp-learning algorithm that allows robots with multi-fingered hands to dexterously grasp previously unseen objects when trained entirely in simulation.

The new approach enabled the robot to grasp with higher than 93% real-world success on novel objects compared to 78% of existing passive learning approaches.

The vacuum and household company, Dyson, is researching household robots.

Giant.AI is working on an upper torso factory bot that can identify objects and pick them up.

SOURCES- DARPA, Oregon State, Giant.AI, Utah, Dyson


Written by Brian Wang, Nextbigfuture.com

Read More

Total
0
Shares
Leave a Reply

Your email address will not be published.

Related Posts
Theo Triantafyllidis Presents: Ork Haus
Read More

Theo Triantafyllidis Presents: Ork Haus

Theo Triantafyllidis presents a nightmarish vision of the metaverse that is part live simulation, part experimental theatre, drawing as much from The Sims as it does Lars Von Trier’s Dogville. “Imagine if you could be at the office without the commute,” enthused Mark Zuckerberg in the 2021 keynote announcing his company’s leap into the metaverse.…
Intonal Festival 2022: Hiro Kone
Read More

Intonal Festival 2022: Hiro Kone

The world premiere of Hiro Kone’s new live show, an extension of her recent album Silvercoat the throng. In the last of a series of highlights from Malmö’s Intonal Festival, we present the world premiere of Disruption and Epokhē, a new live show from New York-based musician and producer Nicky Mao, otherwise known as Hiro Kone.…