Scene understanding device

• US 10,262,215 B2
• Filed: 04/23/2015
• Issued: 04/16/2019
• Est. Priority Date: 04/23/2015
• Status: Active Grant

• Current Assignee: Nissan Motor Co., Ltd.
• Original Assignee: Nissan Motor Co., Ltd.
• Inventors: Yoshihira, Masanori, Watanabe, Seigo, Kishi, Norimasa
• Primary Examiner(s): Tran, Khoi H
• Assistant Examiner(s): Nguyen, Robert T
• Application Number:
  US15/568,302
• Publication Number:
  US 20180144201A1
• Time in Patent Office: 1,454 Days
• CPC Class Codes: B60R 21/00 Arrangements or fittings on vehicles for protecting or preventing injuries to occupants or pedestrians in case of accidents or other traffic risks safety belts or body harnesses in vehicles B60R22/00; devices, apparatus or methods for life-saving in general A62B; safety devices for propulsion unit control specially adapted for, or arranged in, vehicles B60K28/00; seats constructed to protect the occupant from the effect of abnormal g-forces, e.g. crash or safety seats, B60N2/42; energy-absorbing arrangements for hand wheels for steering vehicles B62D1/11; energy-absorbing arrangements for vehicle steering columns B62D1/19; harnessing in aircraft B64D25/00
• CPC Class Codes: B60W 30/095 Predicting travel path or likelihood of collision
• CPC Class Codes: G06T 2207/30261 Obstacle
• CPC Class Codes: G06T 7/70 Determining position or orientation of objects or cameras camera calibration G06T7/80
• CPC Class Codes: G06V 20/58 Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
• CPC Class Codes: G06V 20/588 Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
• CPC Class Codes: G08G 1/16 Anti-collision systems road vehicle drive control systems for predicting or avoiding probable or impending collision otherwise than by control of a particular sub-unit B60W30/08
• CPC Class Codes: G08G 1/165 for passive traffic, e.g. including static obstacles, trees
• CPC Class Codes: G08G 1/166 for active traffic, e.g. moving vehicles, pedestrians, bikes

Abstract

A scene understanding device obtains map data where one, two or more obstacle detection frames, shaped corresponding to a road structure, for detecting an obstacle are set in advance for a specific spot on a road where a vehicle would otherwise bump into a vehicle or a pedestrian. The scene understanding device determines whether there exists an obstacle in the obstacle detection frames set for the specific spot on a scheduled traveling route of a host vehicle, and calculates the degree of risk at the specific spot based on a result of determining whether there exists an obstacle therein.

Claims

1. A scene understanding device that determines a degree of risk at a specific spot where a vehicle would otherwise bump into a vehicle or a pedestrian caused by crossed roads, comprising a computer programmed to:

obtain map data, on a scheduled traveling route, of a plurality of obstacle detection frames for detecting an obstacle, the plurality of obstacle detection frames set for both of the crossed roads in advance;

obtain route data on the scheduled traveling route of a host vehicle;

select a part of the obstacle detection frames to be determined depending on the scheduled traveling route from the plurality of obstacle detection frames set for the specific spot in advance;

determine whether there exists an obstacle in the selected part of the obstacle detection frames; and

calculate the degree of risk at the specific spot based on a result of determining whether an obstacle exists there.

2. The scene understanding device according to claim 1, wherein the computer is further programmed to:

obtain as additional map data a close observation frame which closely observes the specific spot, the close observation frame set for the specific spot in advance,

determine whether a blind spot caused by the obstacle overlaps the close observation frame, and

calculate the degree of risk at the specific spot based on a result of determining whether the blind spot overlaps the close observation frame.

3. The scene understanding device according to claim 1, wherein the plurality of obstacle detection frames is set for the specific spot, and the computer is further programmed to calculate the degree of risk at the specific spot based on a combination of obstacle detection frames including the obstacles, wherein the combination of obstacle detection frames including the obstacles is from the selected part of the obstacle detection frames.

4. The scene understanding device according to claim 3, further comprising:

a degree-of-risk database where the combination of obstacle detection frames including the obstacles is encoded, the computer programmed to store a relationship between the encoded combination and the degree of risk in the degree-of-risk database; and

an encoding processor that encodes the combination of obstacle detection frames including the obstacles, wherein the computer is further programmed to calculate the degree of risk at the specific spot from the encoded combination using the degree-of-risk database.

5. The scene understanding device according to claim 4, wherein the computer is further programmed to detect attributes of the obstacles included in the combination of obstacle detection frames including the obstacles at the specific spot on the scheduled traveling route, and

the encoding processor encodes the combination of obstacle detection frames including the obstacles with the attributes of the obstacles.

6. The scene understanding device according to claim 1, further comprising:

a knowledge database that stores the plurality of obstacle detection frames to be determined depending on a position of the host vehicle at the specific spot, and an order of the plurality of obstacle detection frames to be cautious about; and

a sensor that detects the position of the host vehicle, wherein

referring to the knowledge database, the computer is programmed to determine whether there exists an obstacle in the selected part of the obstacle detection frames sequentially, based on the position of the host vehicle.

7. The scene understanding device according to claim 1, wherein

the specific spot is an intersection where three or more roads meet, and

the plurality of obstacle detection frames are set for an entrance to, an exit from, and an inside of, the intersection, and a pedestrian crossing.