您现在的位置是:UC Berkeley robot navigation could chart a new course for self >>正文
UC Berkeley robot navigation could chart a new course for self
后花园论坛社区|2024夜上海论坛网|爱上海419论坛 -- Back garden77人已围观
简介Using an image of a destination as reference, in this case, the house at the other side of the field...
Using an image of a destination as reference, in this case, the house at the other side of the field, and "priors" about how to navigate overland on various kinds of terrain, gathered from hours of filmed video, an unmanned vehicle navigates nearly two miles on its own to reach its goal.
Shah et al. 2022Robots and self-driving cars have one very large challenge in common, how to navigate the world. Typically, that task is approached by artificial intelligence as a problem of how to map the surroundings, to construct a precise overview of the geometry of a scene before a robot or a car moves across that terrain.
There may be a simpler way.
Electric Vehicles
- The 5 best electric cars: Plus, the cheapest EV available
- The top home EV chargers: Easily fuel up your car
- EV ownership is neither fairytale nor nightmare
- Our favorite electric lawn mowers: The best alternatives to gas-powered
In a paper posted on arXiv Wednesday by scholars at the University of California at Berkeley, a wheeled robot is able to travel kilometers over suburban terrain.
The droid sticks to paths and dodges previously unseen obstacles. Essential is that it doesn't map its environment, as some other approaches have done, such as in autonomous driving AI programs.
Instead, it relies on heuristics picked up from thirty hours of video of previous runs and some overhead maps of the terrain to create an improved schematic of the way stations along the way relate to one another, without a full map.
The research, titled "ViKiNG: Vision-Based Kilometer-Scale Navigation with Geographic Hints," is authored by Ph.D. candidate Dhruv Shah and UC Berkeley assistant professor Sergey Levine.
The ViKiNG systems approach outline. Sampling from images, along with hints, allows the system to build a graph of the local topology on the fly to chart a course to the destination.
Shah et al. 2022The ViKiNG program has increased its training data of camera observations from the random walk with an additional 12 hours of video that was from "teleoperated" trips, where a human guided the Jackal to pursue paths, such as sidewalks or hiking trails, to build up those prior examples. The neural network utilized to crunch all that training data is fairly humdrum, the familiar MobileNet convolutional neural network.
Artificial Intelligence
- AI in 2023: A year of breakthroughs that left no human thing unchanged
- These are the jobs most likely to be taken over by AI
- AI at the edge: 5G and the Internet of Things see fast times ahead
- Almost half of tech executives say their organizations aren't ready for AI or other advanced initiatives
This time, equipped with ViKiNG, Jackal goes well beyond the 80 meters of RECON, traveling from start to destination almost 3 kilometers away, or almost two miles.
In videos featured on a blog page for the project, Shah and Levine show how the Jackal with ViKiNG can figure out how to route around previously unknown obstacles, such as a parked vehicle blocking its path. A companion video explains the work, which you can view at the bottom of this post.
RECON explicitly employed elements of reinforcement learning. ViKiNG, likewise, borrows in some way. Asked about the connection to RL, Levine told ZDNetin an email, "I would characterize ViKiNG as a kind of reinforcement learning method with a higher level planner sitting on top of it."
This AI-powered prosthetic hand is bringing design and style to a life-changing product The best robot vacuums available now Why do college kids get all the cool robots? The 5 best robot mowers: Hands-free lawn care This AI-powered prosthetic hand is bringing design and style to a life-changing productTags:
转载:欢迎各位朋友分享到网络,但转载请说明文章出处“后花园论坛社区|2024夜上海论坛网|爱上海419论坛 -- Back garden”。http://dzbz.com.cn/news/1696.html
相关文章
Math and business development: Why the 2 divergents skills matter in an AI world
UC Berkeley robot navigation could chart a new course for selfDmitriy Sidor/Getty ImagesArtificial intelligence (AI) keeps upending our ideas about future skill r...
阅读更多
Singapore wants to plug AI skills gap in finance sector
UC Berkeley robot navigation could chart a new course for selfSingapore Business District.Getty Images/MR.Cole_PhotographerSingapore is looking to plug a dearth o...
阅读更多
Intel advances oneAPI as the all
UC Berkeley robot navigation could chart a new course for selfIntelThe rise of "generative" artificial intelligence is all about scaling, the idea of adding more...
阅读更多
热门文章
- Top of the data heap: First
- AI projects now exceed 350,000, according to Stanford
- IT leaders aren't getting listened to, and now they're ready to walk away
- Stack Overflow joins Reddit and Twitter in charging AI companies for training data
- Check your SPF records: Wide IP ranges undo email security and make for tasty phishes
- Google's new tool can detect AI
最新文章
As developers learn the ins and outs of generative AI, non
The metaverse has virtually disappeared. Here's why it's generative AI's fault
Can generative AI solve computer science's greatest unsolved problem?
Microsoft is testing an AI
Facebook will just create your stories for you now
Microsoft has over a million paying Github Copilot users: CEO Nadella