US 12,299,899 B2
Systems and methods for real-time state estimation of fast-moving objects
Ziyun Wang, Mountain View, CA (US); Fernando Cladera Ojeda, Philadelphia, PA (US); Anthony Robert Bisulco, New York, NY (US); Dae Won Lee, Princeton, NJ (US); Camillo J. Taylor, Mountain View, CA (US); Konstantinos Daniilidis, Wynnewood, PA (US); Ani Hsieh, Marlton, NJ (US); and Ibrahim Volkan Isler, Saint Paul, MN (US)
Assigned to SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR); and The Trustees of the University of Pennsylvania, Philadelphia, PA (US)
Filed by SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR); and The Trustees of the University of Pennsylvania, Philadelphia, PA (US)
Filed on Nov. 1, 2022, as Appl. No. 17/978,873.
Claims priority of provisional application 63/274,739, filed on Nov. 2, 2021.
Prior Publication US 2023/0136306 A1, May 4, 2023
Int. Cl. G06T 7/20 (2017.01); A63B 65/12 (2006.01); G06V 10/82 (2022.01); G06V 20/40 (2022.01)
CPC G06T 7/20 (2013.01) [A63B 65/12 (2013.01); G06V 10/82 (2022.01); G06V 20/44 (2022.01); G06T 2207/30241 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for predicting a location of a fast-moving object, the method comprising:
receiving event information from an event camera, the event information corresponding to an event detected by the event camera;
generating a Binary Event History Image (BEHI) based on the event information;
providing the BEHI as an input to an event-based neural network;
in response to inputting the BEHI, obtaining, as an output of the event-based neural network, a first predicted location of the fast-moving object, a normal distribution indicating prediction uncertainty of the predicted location, and a predicted time-to-collision (TTC);
estimating a second predicted location of the fast-moving object based on the first predicted location, the normal distribution, and the predicted TTC output by the event-based neural network; and
actuating a mechanical catching device to be at the second predicted location.