3 ms·
Nutonomy founder Karl gave a talk at MIT and one of the things that surprised me was that they used very little machine learning. Most the work is done using a
by anilshanbhag 9y ago
Nutonomy founder Karl gave a talk at MIT and one of the things that surprised me was that they used very little machine learning. Most the work is done using a pre defined route map and the rrt star algorithm for making local decisions.
- isatty 9y agoThat's actually a good thing - I've done a fair amount of work on autonomous robots (not land) and I'd say that the AI/Machine Learning word is thrown about a lot for no reason. Many of their challenges can be solved with tried and proven algorithms like rrt star. My personal opinion is that they're a autonomous car company not a research lab, so the focus is indeed correct.
- kajecounterhack 9y agoMost teams use ml heavily for perception but not necessarily for other things. You need explainability when it comes to safety, and maps are needed no matter what if you want your car to get around.
- bobsil1 9y agoThe opposite of Drive.ai and Nvidia approaches (end-to-end deep learning).
- gumby 9y agoThat's true of all the autonomous vehicle projects: the planners, etc are all written by hand. The ML is for "image" processing (I say "image" because it's object extraction from multiple sensors, in most cases including LIDAR).
- sjg007 9y agoand those outputs probably map into a bayesian network to make decisions.