4 ms·
Interestingly, for air travel, there's another strategy: Rely entirely on GPS and commanding from air traffic control (with systems and structure in place to mi
by Robotbeat 8y ago
Interestingly, for air travel, there's another strategy:
Rely entirely on GPS and commanding from air traffic control (with systems and structure in place to minimize damage in case those systems fail or are insufficient).
That's what Zipline uses for their fully operational medical delivery system in Rwanda.
The drones are completely blind, directly, but they travel in pre-defined flight paths and communicate with one another and follow directions from flight control.
For self-driving cars, a similar situation may be extremely good GPS geolocation combined with network-wide sensor fusion and mapping. In principle, you could even do real-time optical or synthetic aperture radar from orbit or via persistent aircraft to allow the central controller to identify hazards and obstacles and to update maps in real time without much at all happening on-board except low-latency reactions. Most highways already have much of their length covered by cameras for the local Department of Transportation, and even many intersections and sidestreets have surveillance cameras. A low latency connection to that system, upgraded with higher fidelity, could significantly help autonomous systems. Might be another very helpful public good for cities to provide, much like GPS is provided.
...and on the engineering side, you can have vehicles designed specifically to reduce pedestrian injury, such as not having those large, boxy grilles on SUV. Such things are almost entirely cosmetic and reduce efficiency, so Tesla doesn't have them on their cars. Going beyond that, external airbags may help as well. Zipline uses a foam body and a fail-safe parachute to stop the vehicle if there's a problem.
- kec 8y agoIf a plane can only get a GPS fix to within a 15m radius that's no big deal, whereas if a car has a 15m fix it's running down pedestrians on the sidewalk. Planes also don't have a habit of flying in tunnels or urban canyons where obtaining a gps fix is even more problematic.
- Robotbeat 8y agoIndeed. But this is an addressable problem. The new GPS satellites (GPS III) have just started launching (on a SpaceX rocket in December 2018), and they should triple the accuracy: https://www.geospatialworld.net/blogs/what-is-gps-iii-and-how-will-it-stand-apart/ https://www.geospatialworld.net/blogs/what-is-gps-iii-and-ho... But even before that's complete: High precision GPS (i.e. with compensation for atmospheric effects) can get better than 10cm accuracy by tracking the carrier wave: https://en.wikipedia.org/wiki/Real-time_kinematic https://en.wikipedia.org/wiki/Real-time_kinematic Single-digit-centimeter-accuracy is possible near GPS signal compensating base stations (~20km). This level of precision is regularly used by surveyors. There are several options for improving geolocation beyond inside-out (i.e. vision or lidar) tracking.
- lutorm 8y agoPlanes already have a problem with GPS outages or military jamming exercises. Imagine if all cars within a 100 mile radius all of a sudden would stop working.
- Robotbeat 8y ago...that's why you develop systems designed to bring them to a safe stop (in addition, stable IMUs can assist for several seconds, plenty of time to come to a safe stop). That should be standard on all autonomous systems, TBH. And BTW, there are 4 GNSS systems, each run by a different country/entity. The odds of them all failing at the same time is very small.
- ksherlock 8y agoGPS is (negatively) affected by atmospheric variability. Are the other 3 systems immune from it?