PDF Link attached. Abridged summary from The Register here: Remember the Uber self-driving car that killed a woman crossing the street? The AI had no clue about jaywalkers
PDF Link attached. Abridged summary from The Register here:
Seems like the autonomous program plays its role much too aggressively. As this kind of technology is still in it's infancy, it's absolutely acceptable for it to err on the side of caution. Aside...
Although the ADS sensed the pedestrian nearly 6 seconds before the impact, the system never classified her as a pedestrian—or predicted correctly her goal as a jaywalking pedestrian or a cyclist—because she was crossing the N. Mill Avenue at a location without a crosswalk; the system design did not include a consideration for jaywalking pedestrians. Instead, the system had initially classified her as an other object which are not assigned goals. As the ADS changed the classification of the pedestrian several times—alternating between vehicle, bicycle, and an other—the system was unable to correctly predict the path of the detected object.
Only when the ADS determined that the object’s currently detected location was on the path of the ATG vehicle—1.2 seconds before impact—the system recognized an emergency situation, an imminent collision. At that time, because preventing the collision would have required extreme braking or steering actions—beyond the design specifications—the ADS initiated suppression of its motion plan. One second later, the vehicle was still on the collision path with the pedestrian, and preventing the collision still required an extreme avoidance maneuverer; per design, the system did not engage emergency brakes, but rather provided an auditory alert to thevehicle operator as it initiated a plan for the vehicle slowdown.
Seems like the autonomous program plays its role much too aggressively. As this kind of technology is still in it's infancy, it's absolutely acceptable for it to err on the side of caution.
Aside from the odd choice to only identify pedestrians at pre-positioned crosswalks, there is no reason for the SUV not to at least slow down (or hand over control) for unidentified objects just a few seconds from its path. Sure, 95% of the time it may be a misidentified stopped car or rock, but there should be little doubt before the computer hurls its 4,000 pounds of steel and aluminum next to a smaller moving object. At this point in time, experimental autonomous vehicle development should focus on safety and not making the ride less interruption- or intervention-prone. Hopefully this report helps with future autonomous vehicle development. Definitely would like to see the day where this technology makes the road safer for all drivers.
PDF Link attached. Abridged summary from The Register here:
Seems like the autonomous program plays its role much too aggressively. As this kind of technology is still in it's infancy, it's absolutely acceptable for it to err on the side of caution.
Aside from the odd choice to only identify pedestrians at pre-positioned crosswalks, there is no reason for the SUV not to at least slow down (or hand over control) for unidentified objects just a few seconds from its path. Sure, 95% of the time it may be a misidentified stopped car or rock, but there should be little doubt before the computer hurls its 4,000 pounds of steel and aluminum next to a smaller moving object. At this point in time, experimental autonomous vehicle development should focus on safety and not making the ride less interruption- or intervention-prone. Hopefully this report helps with future autonomous vehicle development. Definitely would like to see the day where this technology makes the road safer for all drivers.