# US 2019/0171896 A1 · Reliability calculation apparatus

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2019/0171896 A1
- Document ID: US20190171896A1
- Status: Published application
- Kind: A1
- Application number: 16/206,004
- Filed: 2018-11-30
- Published: 2019-06-06
- Earliest claimed: 2017-12-05
- Assignees: Denso Corporation
- Applicants: Denso Corporation
- Inventors: Masaya OKADA; Shunya KUMANO; Hiroaki NIINO; Mitsuhiro TOKIMASA; Takumi UEMATSU
- Web page: https://patentgrep.com/patent/US20190171896A1

## CPC

- G06V 20/588: {Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road}
- G06F 18/22: Matching criteria, e.g. proximity measures
- G06T 7/50: Depth or shape recovery
- G06V 10/255: {Detecting or recognising potential candidate objects based on visual cues, e.g. shapes}
- G06V 10/98: Detection or correction of errors, e.g. by rescanning the pattern or by human intervention; Evaluation of the quality of the acquired patterns
- 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
- G06T 2207/30256: Lane; Road marking
- G06V 2201/08: Detecting or categorising vehicles
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A reliability calculation apparatus calculates a reliability of a recognized object that is recognized as being present on a route on which a vehicle travels. The reliability of the recognized object is used for driving control of the vehicle. In the reliability calculation apparatus, a provisional reliability calculating unit determines a feature quantity of each of a plurality of recognized objects, selects a plurality of pairs of recognized objects from the plurality of recognized objects, determines a difference in feature quantity of each pair of recognized objects, and calculates a reliability that decreases as the difference in feature quantity increases as a provisional reliability of each pair of recognized objects. An integrated reliability calculating unit calculates an integrated reliability of each recognized object from the provisional reliabilities of the plurality of pairs of recognized objects.

## Claims

### Claim 1

A reliability calculation apparatus that calculates a reliability of a recognized object that is recognized as being present on a route on which a vehicle travels, the reliability of the recognized object being used for driving control of the vehicle, the reliability calculation apparatus comprising:
a provisional reliability calculating unit that determines a feature quantity of each of a plurality of recognized objects, selects a plurality of pairs of recognized objects from the plurality of recognized objects, determines a difference in feature quantity of each pair of recognized objects, and calculates, as a provisional reliability of each pair of recognized objects, a reliability that decreases as the difference in feature quantity increases; and
an integrated reliability calculating unit that calculates an integrated reliability of each recognized object from the provisional reliabilities of the plurality of pairs of recognized objects.

### Claim 2 (depends on 1)

The reliability calculation apparatus according to claim 1, wherein:
the integrated reliability calculating unit calculates the integrated reliability of each recognized object using a predetermined function in which the provisional reliability is used as a variable.

### Claim 3 (depends on 1)

The reliability calculation apparatus according to claim 1, wherein:
the plurality of recognized objects include at least one of a plurality of types including travel road boundary lines, a travel locus of another vehicle, and roadside objects; and
the provisional reliability calculating unit calculates a shape parameter that indicates a shape of the recognized object as the feature quantity.

### Claim 4 (depends on 1)

The reliability calculation apparatus according to claim 1, wherein:
the plurality of recognized objects include a travel road boundary line and a roadside object; and
the provisional reliability calculating unit sets the provisional reliability related to a pair of recognized objects of the travel road boundary line and the road side object to a value that is lower than a predetermined value when the travel road boundary line is positioned further towards an outer side than the roadside object.

### Claim 5 (depends on 1)

The reliability calculation apparatus according to claim 1, wherein:
the plurality of recognized objects include a travel road boundary line and a roadside object; and
the provisional reliability calculating unit sets the provisional reliability related to a pair of recognized objects of the travel road boundary line and the road side object when the travel road boundary line is positioned further towards an outer side than the roadside object to a value that is lower than a value when the travel road boundary line is positioned further towards an inner side than the roadside object is.

### Claim 6 (depends on 1)

The reliability calculation apparatus according to claim 1, further comprising:
an individual reliability calculating unit that calculates an individual reliability of each recognized object by a method differing from that of the provisional reliability calculating unit, wherein
the integrated reliability calculating unit calculates the integrated reliability of each recognized object from the provisional reliability and the individual reliability.

### Claim 7 (depends on 3)

The reliability calculation apparatus according to claim 3, wherein:
when a predetermined extrapolation condition that includes a condition that, of a first recognized object and a second recognized object composing the pair of recognized objects, the first recognized object reaches farther than the second recognized object is met, the provisional reliability calculating unit extrapolates the second recognized object to a distance equal to a distance reached by the first recognized object using an approximate curve of the second recognized object, calculates the shape parameter of the second recognized object, and determines a difference in shape parameter between the first recognized object and the second recognized object.

### Claim 8 (depends on 7)

The reliability calculation apparatus according to claim 7, wherein:
the integrated reliability calculating unit uses a function that is configured such that the provisional reliability does not affect a value of the integrated reliability when the provisional reliability is set to a specific value, as the function for calculating the integrated reliability; and
when the second recognized object is extrapolated, and the difference in shape parameter between the first recognized object and the second recognized object is greater than a predetermined difference threshold, the provisional reliability calculating unit sets the provisional reliability to become closer to the specific value when a distance of the extrapolation is long, compared to when the distance is short.

### Claim 9 (depends on 7)

The reliability calculation apparatus according to claim 7, wherein:
the provisional reliability calculating unit does not calculate the provisional reliability when the second recognized object is extrapolated and an extrapolation distance exceeds a predetermined upper-limit value.

### Claim 10 (depends on 7)

The reliability calculation apparatus according to claim 7, wherein:
the integrated reliability calculating unit uses a function that is configured such that the provisional reliability does not affect a value of the integrated reliability when the provisional reliability is set to a specific value, as the function for calculating the integrated reliability; and
when the second recognized object is extrapolated, and the difference in shape parameter between the first recognized object and the second recognized object is greater than a predetermined difference threshold, the provisional reliability calculating unit sets the provisional reliability to become closer to the specific value when a recognition distance of the second recognized object is short, compared to when the recognition distance is long.

### Claim 11 (depends on 10)

The reliability calculation apparatus according to claim 10, wherein:
when the second recognized object is extrapolated and the recognition distance of the second recognized object is less than a predetermined lower-limit value, the provisional reliability calculating unit does not calculate the provisional reliability.

### Claim 12 (depends on 7)

The reliability calculation apparatus according to claim 7, wherein:
the extrapolation condition further includes a condition that the provisional reliability used to calculate the integrated reliability of the first recognized object is being calculated.

### Claim 13 (depends on 7)

The reliability calculation apparatus according to claim 7, wherein:
the extrapolation condition further includes meeting of either of:
(1) the provisional reliability used to calculate the integrated reliability of the first recognized object being calculated; and
(2) the provisional reliability used to calculate the integrated reliability of the second recognized object is being calculated, and a difference between the shape parameter at a far end of the second recognized object and the shape parameter at a far end of the first recognized object being equal to or greater than a predetermined difference upper-limit value.

### Claim 14 (depends on 2)

The reliability calculation apparatus according to claim 2, wherein:
the plurality of recognized objects include at least one of a plurality of types including travel road boundary lines, a travel locus of another vehicle, and roadside objects; and
the provisional reliability calculating unit calculates a shape parameter that indicates a shape of the recognized object as the feature quantity.

### Claim 15 (depends on 2)

The reliability calculation apparatus according to claim 2, wherein:
the plurality of recognized objects include a travel road boundary line and a roadside object; and
the provisional reliability calculating unit sets the provisional reliability related to a pair of recognized objects of the travel road boundary line and the road side object to a value that is lower than a predetermined value when the travel road boundary line is positioned further towards an outer side than the roadside object is.

### Claim 16 (depends on 3)

The reliability calculation apparatus according to claim 3, wherein:
the plurality of recognized objects include a travel road boundary line and a roadside object; and
the provisional reliability calculating unit sets the provisional reliability related to a pair of recognized objects of the travel road boundary line and the road side object to a value that is lower than a predetermined value when the travel road boundary line is positioned further towards an outer side than the roadside object is.

### Claim 17 (depends on 2)

The reliability calculation apparatus according to claim 2, wherein:
the plurality of recognized objects include a travel road boundary line and a roadside object; and
the provisional reliability calculating unit sets the provisional reliability related to a pair of recognized objects of the travel road boundary line and the road side object when the travel road boundary line is positioned further towards an outer side than the roadside object to a value that is lower than a value when the travel road boundary line is positioned further towards an inner side than the roadside object is.

### Claim 18 (depends on 3)

The reliability calculation apparatus according to claim 3, wherein:
the plurality of recognized objects include a travel road boundary line and a roadside object; and
the provisional reliability calculating unit sets the provisional reliability related to a pair of recognized objects of the travel road boundary line and the road side object when the travel road boundary line is positioned further towards an outer side than the roadside object is to a value that is lower than a value when the travel road boundary line is positioned further towards an inner side than the roadside object is.

### Claim 19

A vehicle control system comprising:
a reliability calculation apparatus that calculates a reliability of a recognized object that is recognized as being present on a route on which a vehicle travels; and
a control apparatus that performs driving control of the vehicle based on the reliability of the recognized object calculated by the reliability calculation apparatus,
the reliability calculation apparatus comprising:
a provisional reliability calculating unit that determines a feature quantity of each of a plurality of recognized objects, selects a plurality of pairs of recognized objects from the plurality of recognized objects, determines a difference in feature quantity of each pair of recognized objects, and calculates, as a provisional reliability of each pair of recognized objects, a reliability that decreases as the difference in feature quantity increases; and
an integrated reliability calculating unit that calculates an integrated reliability of each recognized object from the provisional reliabilities of the plurality of pairs of recognized objects.

### Claim 20

A reliability calculation method for calculating a reliability of a recognized object that is recognized as being present on a route on which a vehicle travels, the reliability of the recognized object being used for driving control of the vehicle, the reliability calculation method comprising:
determining a feature quantity of each of a plurality of recognized objects;
selecting a plurality of pairs of recognized objects from the plurality of recognized objects;
determining a difference in feature quantity of each pair of recognized objects;
calculating, as a provisional reliability of each pair of recognized objects, a reliability that decreases as the difference in feature quantity increases; and
calculating an integrated reliability of each recognized object from the provisional reliabilities of the plurality of pairs of recognized objects.
