US 11,776,396 B2
Intersection infrastructure warning system
Rajesh Kumar Malhan, Troy, MI (US); Joseph Lull, South Haven, MI (US); and Tishya Bhatia, Ypsilanti, MI (US)
Assigned to DENSO International America, Inc., Southfield, MI (US)
Filed by DENSO International America, Inc., Southfield, MI (US)
Filed on Mar. 31, 2020, as Appl. No. 16/836,221.
Claims priority of provisional application 62/949,022, filed on Dec. 17, 2019.
Prior Publication US 2021/0183244 A1, Jun. 17, 2021
Int. Cl. G08G 1/09 (2006.01); G08G 1/005 (2006.01); G06T 1/20 (2006.01); B64D 47/02 (2006.01); B64D 47/08 (2006.01); B64F 1/36 (2017.01); B64C 39/02 (2023.01); G06V 20/10 (2022.01); B64U 10/13 (2023.01); B64U 101/00 (2023.01)
CPC G08G 1/09 (2013.01) [B64C 39/024 (2013.01); B64D 47/02 (2013.01); B64D 47/08 (2013.01); B64F 1/362 (2013.01); G06T 1/20 (2013.01); G06V 20/10 (2022.01); G08G 1/005 (2013.01); B64U 10/13 (2023.01); B64U 2101/00 (2023.01)] 20 Claims
OG exemplary drawing
 
1. An intersection infrastructure warning system for warning vulnerable road users (VRUs) of danger around an intersection, the warning system comprising:
an edge computing device having a graphics processing unit (GPU), a central processing unit (CPU), a storage device, and a communications module configured for wired or wireless communication, the edge computing device configured to communicate with a distributed cloud networking system;
the edge computing device located at the intersection;
one or more sensors configured to acquire spatial-temporal data of road users, including the VRUs, in a 360° area around the intersection; and
one or more warning devices configured to output a warning to warn targeted VRUs of the danger around the intersection, wherein
the GPU is configured to use one or more artificial intelligence (AI) algorithms to process the spatial-temporal data of the road users to determine predictions of the road users, the predictions including at least one of a path, a trajectory, a behavior, and an intent of a road user,
the GPU is further configured to analyze the predictions of the road users to determine convergences between the predictions of the road users to determine predicted collisions and identify the targeted VRUs,
in response to determining predicted collisions and identifying the targeted VRUs, the edge computing device is further configured to output targeting instructions and warning response instructions to the one or more warning devices to deploy the one or more warning devices to the targeted VRUs, and
in response to receiving the targeting instructions and warning response instructions, the one or more warning devices are further configured to target the targeted VRUs based on the targeting instructions and to output the warning to the targeted VRUs; and
wherein the VRUs and the targeted VRUs are nonmotorists.