Vehicle position estimation device

Abstract

The present disclosure relates to a vehicle position estimation device. The vehicle position estimation device (100) for estimating the position of a vehicle comprises: a laser irradiation unit (11) for irradiating laser light; a light receiving element (12) for receiving reflected light of the laser light and reflected light of ambient light other than the laser light, and detecting the intensity of each received reflected light; and a self-position estimation unit (21) for estimating the position of the vehicle based on the light reception result of the reflected light of the laser light and the light reception result of the reflected light of the ambient light received by the light receiving element (12).

Description

Technical Field

[0001] The present disclosure relates to a vehicle position estimating device. Background Art

[0002] For example, in order to detect surrounding target objects, a laser radar (LIDAR) is mounted on the vehicle to detect the target objects based on the reflected light of the irradiated laser. In addition, as described in non-patent document 1, a laser radar is developed that detects the target objects by using the light reception result of the reflected light of the ambient light other than the irradiated laser in addition to the reflected light of the irradiated laser.

[0003] Prior art literature

[0004] Non-patent literature

[0005] Non-patent document 1: Seigo Ito, Seira Hiratsuka, Mitsuhiko Matsubara, Hiroyuki Ogawa, "Small LIDAR DCNN Position and Posture Estimation" Lecture Papers at the 79th National Conference of the Information Processing Society Summary of the invention

[0006] Problems to be solved by the invention

[0007] The position of the vehicle is estimated by matching the detection results of the laser radar mounted on the vehicle with the position information of the object included in the map information. In this case, for example, there is a situation where it is difficult to estimate the position of the vehicle when using an object that is difficult to detect with a laser. As described above, a laser radar that also uses the light reception result of the reflected light of the ambient light is disclosed as a detection device for detecting the object, but there is no disclosure of applying such a laser radar to the above situation for the position estimation of the vehicle, and there is room for research.

[0008] Therefore, the present disclosure describes a vehicle position estimating device that can estimate the position of the host vehicle with high accuracy using reflected light of ambient light other than laser light in addition to reflected light of laser light.

[0009] Technical solutions to solve problems

[0010] A technical solution disclosed in the present invention is a vehicle position estimating device for estimating the position of the vehicle, comprising: a laser irradiation unit for irradiating laser; a light receiving unit for receiving reflected light of the laser and reflected light of ambient light which is light other than the laser, and detecting the intensity of each reflected light received; and an estimating unit for estimating the position of the vehicle based on the light reception results of the reflected light of the laser and the reflected light of the ambient light received by the light receiving unit.

[0011] The vehicle position estimation device includes an estimation unit that estimates the position of the vehicle based on the light reception result of the reflected light of the laser and the light reception result of the reflected light of the ambient light. For example, even when there is an object that is difficult to detect by the laser, the object can sometimes be detected based on the reflected light of the ambient light. Therefore, the vehicle position estimation device can estimate the position of the vehicle with good accuracy by using the reflected light of the ambient light other than the laser in addition to the reflected light of the laser.

[0012] The vehicle position estimation device may also include a map storage unit, which pre-stores map laser information based on the light reception result of the reflected light of the laser and map ambient light information based on the light reception result of the reflected light of the ambient light for each location on the map, and the estimation unit calculates laser matching information and ambient light matching information, the laser matching information being the matching result of the laser information based on the light reception result of the reflected light of the laser received by the light receiving unit and the map laser information, and the ambient light matching information being the matching result of the ambient light information based on the light reception result of the reflected light of the ambient light received by the light receiving unit and the map ambient light information, and estimates the position of the vehicle based on the calculated laser matching information and ambient light matching information. In this way, the vehicle position estimation device first calculates the laser matching information and the ambient light matching information, and then estimates the position of the vehicle based on the laser matching information and the ambient light matching information. In this case, the vehicle position estimation device can use the laser matching information and the ambient light matching information to accurately estimate the position of the vehicle.

[0013] The vehicle position estimation device may also include a map storage unit, which stores map laser information based on the light reception result of the reflected light of the laser and map ambient light information based on the light reception result of the reflected light of the ambient light in advance for each location on the map. The estimation unit calculates sensor comprehensive information and map comprehensive information. The sensor comprehensive information is the comprehensive result of laser information based on the light reception result of the reflected light of the laser received by the light receiving unit and ambient light information based on the light reception result of the reflected light of the ambient light received by the light receiving unit. The map comprehensive information is the comprehensive result of the map laser information and the map ambient light information. The position of the vehicle is estimated based on the calculated sensor comprehensive information and map comprehensive information. In this way, the vehicle position estimation device first calculates the sensor comprehensive information by integrating the laser information based on the light reception result of the light receiving unit and the ambient light information, and calculates the map comprehensive information by integrating the map laser information and the map ambient light information stored in the map storage unit. Next, the vehicle position estimation device estimates the position of the vehicle based on the calculated sensor comprehensive information and map comprehensive information. In this case, the vehicle position estimating device can estimate the position of the host vehicle with high accuracy based on the sensor integrated information and the map integrated information.

[0014] The vehicle position estimation device may also include a map storage unit, which stores map comprehensive information in advance for each location on the map, the map comprehensive information being a comprehensive result of map laser information based on the light reception result of the reflected light of the laser and map ambient light information based on the light reception result of the reflected light of the ambient light, and the estimation unit calculates sensor comprehensive information, which is a comprehensive result of laser information based on the light reception result of the reflected light of the laser received by the light receiving unit and ambient light information based on the light reception result of the reflected light of the ambient light received by the light receiving unit, and estimates the position of the vehicle based on the map comprehensive information and the calculated sensor comprehensive information. In this way, the map storage unit in the vehicle position estimation device stores map comprehensive information obtained by integrating the map laser information based on the light reception result of the light receiving unit and the map ambient light information in advance for each location on the map. In this case, the vehicle position estimation device first calculates the sensor comprehensive information obtained by integrating the laser information based on the light reception result of the light receiving unit and the ambient light information. Then, the vehicle position estimation device estimates the position of the vehicle based on the map comprehensive information and the sensor comprehensive information. In this case, the vehicle position estimation device can estimate the position of the vehicle with good accuracy based on the sensor comprehensive information and the map comprehensive information. In addition, since the map comprehensive information is pre-stored in the map storage unit, the vehicle position estimation device can estimate the position of the vehicle at a higher speed.

[0015] The vehicle position estimation device may also include a map storage unit, which pre-stores map laser information based on the light reception result of the reflected light of the laser and map ambient light information based on the light reception result of the reflected light of the ambient light for each location on the map, the estimation unit extracts a feature quantity for object detection as a laser feature quantity from the light reception result of the reflected light of the laser received by the light receiving unit, and extracts a feature quantity for object detection as a map laser feature quantity from the map laser information, calculates laser feature quantity information, which is a matching result of the extracted laser feature quantity and the map laser feature quantity, extracts a feature quantity for object detection as an ambient light feature quantity from the light reception result of the reflected light of the ambient light received by the light receiving unit, and extracts a feature quantity for object detection as a map ambient light feature quantity from the map ambient light information, calculates ambient light feature quantity information, which is a matching result of the extracted ambient light feature quantity and the map ambient light feature quantity, and estimates the position of the vehicle based on the calculated laser feature quantity information and ambient light feature quantity information. In this way, the vehicle position estimation device estimates the position of the vehicle using the feature quantity used for object detection. Here, the vehicle position estimation device first extracts the laser feature quantity from the light reception result of the laser of the light receiving unit, and extracts the map laser feature quantity from the map laser information, and matches the laser feature quantity with the map laser feature quantity to calculate the laser feature quantity information. In addition, the vehicle position estimation device extracts the ambient light feature quantity from the light reception result of the ambient light of the light receiving unit, and extracts the map ambient light feature quantity from the map ambient light information, and matches the ambient light feature quantity with the map ambient light feature quantity to calculate the ambient light feature quantity information. Next, the vehicle position estimation device estimates the position of the vehicle based on the laser feature quantity information and the ambient light feature quantity information. In this case, the vehicle position estimation device can estimate the position of the vehicle with good accuracy based on the laser feature quantity information and the ambient light feature quantity information.

[0016] The vehicle position estimation device may further include a map storage unit, the map storage unit pre-storing a map laser feature quantity and a map ambient light feature quantity for each location on the map, the map laser feature quantity being a feature quantity for object detection extracted from a result of receiving reflected light of the laser, and the map ambient light feature quantity being a feature quantity for object detection extracted from a result of receiving reflected light of the ambient light, the estimation unit extracting a feature quantity for object detection from a result of receiving reflected light of the laser received by the light receiving unit as a laser feature quantity, calculating laser feature quantity information, the laser feature quantity information being a result of matching between the map laser feature quantity and the extracted laser feature quantity, extracting a feature quantity for object detection from a result of receiving reflected light of the ambient light received by the light receiving unit as an ambient light feature quantity, calculating ambient light feature quantity information, the ambient light feature quantity information being a result of matching between the map ambient light feature quantity and the extracted ambient light feature quantity, and estimating the position of the vehicle based on the calculated laser feature quantity information and the ambient light feature quantity information. In this way, the vehicle position estimation device estimates the position of the vehicle using the feature quantity for object detection. Here, the map storage unit in the vehicle position estimation device pre-stores the map laser feature extracted from the light reception result of the reflected light of the laser and the map ambient light feature extracted from the light reception result of the reflected light of the ambient light for each location on the map. In this case, the vehicle position estimation device first calculates the laser feature information as the matching result of the laser feature extracted from the light reception result of the reflected light of the laser received by the light receiving unit and the map laser feature, and calculates the ambient light feature information as the matching result of the ambient light feature extracted from the light reception result of the reflected light of the ambient light received by the light receiving unit and the map ambient light feature. Next, the vehicle position estimation device estimates the position of the vehicle based on the laser feature information and the ambient light feature information. In this case, the vehicle position estimation device can estimate the position of the vehicle with good accuracy based on the laser feature information and the ambient light feature information. In addition, since the map laser feature and the map ambient light feature are pre-stored in the map storage unit, the vehicle position estimation device can estimate the position of the vehicle at a higher speed.

[0017] The vehicle position estimation device may also include a map storage unit, which pre-stores map laser information based on the light reception result of the reflected light of the laser and map ambient light information based on the light reception result of the reflected light of the ambient light for each location on the map, the estimation unit extracts a feature quantity for object detection as a laser feature quantity from the light reception result of the reflected light of the laser received by the light receiving unit, and extracts a feature quantity for object detection as an ambient light feature quantity from the light reception result of the reflected light of the ambient light received by the light receiving unit, calculates sensor feature quantity information, which is a comprehensive result of the extracted laser feature quantity and the ambient light feature quantity, extracts a feature quantity for object detection from the map laser information as a map laser feature quantity, and extracts a feature quantity for object detection from the map ambient light information as a map ambient light feature quantity, calculates map feature quantity information, which is a comprehensive result of the extracted map laser feature quantity and the map ambient light feature quantity, and estimates the position of the vehicle based on the calculated sensor feature quantity information and map feature quantity information. In this way, the vehicle position estimation device estimates the position of the vehicle using the feature quantity used for object detection. Here, the vehicle position estimation device first extracts the laser feature quantity from the light receiving result of the laser of the light receiving unit, and extracts the ambient light feature quantity from the light receiving result of the ambient light of the light receiving unit, and integrates the extracted laser feature quantity and the ambient light feature quantity to calculate the sensor feature quantity information. In addition, the vehicle position estimation device extracts the map laser feature quantity from the map laser information, and extracts the map ambient light feature quantity from the map ambient light information, and integrates the extracted map laser feature quantity and the map ambient light feature quantity to calculate the map feature quantity information. Next, the vehicle position estimation device estimates the position of the vehicle based on the sensor feature quantity information and the map feature quantity information. In this case, the vehicle position estimation device can estimate the position of the vehicle with good accuracy based on the sensor feature quantity information and the map feature quantity information.

[0018] The vehicle position estimation device may also include a map storage unit, the map storage unit pre-storing map feature quantity information for each location on the map, the map feature quantity information being a comprehensive result of a map laser feature quantity and a map ambient light feature quantity, the map laser feature quantity being a feature quantity for object detection extracted from a light reception result of reflected light of the laser, and the map ambient light feature quantity being a feature quantity for object detection extracted from a light reception result of reflected light of ambient light, the estimation unit extracting a feature quantity for object detection from a light reception result of reflected light of the laser received by the light receiving unit as a laser feature quantity, and extracting a feature quantity for object detection from a light reception result of reflected light of ambient light received by the light receiving unit as an ambient light feature quantity, calculating sensor feature quantity information, the sensor feature quantity information being a comprehensive result of the extracted laser feature quantity and the ambient light feature quantity, and estimating the position of the vehicle based on the map feature quantity and the calculated sensor feature quantity information. In this way, the vehicle position estimation device estimates the position of the vehicle using the feature quantity for object detection. Here, the map storage unit of the vehicle position estimation device pre-stores map feature information obtained by integrating the map laser feature extracted from the light reception result of the reflected light of the laser and the map ambient light feature extracted from the light reception result of the reflected light of the ambient light for each location on the map. In this case, the vehicle position estimation device first extracts the laser feature from the light reception result of the reflected light of the laser received by the light receiving unit, and extracts the ambient light feature from the light reception result of the reflected light of the ambient light received by the light receiving unit, and calculates the sensor feature information obtained by integrating the laser feature and the ambient light feature. Then, the vehicle position estimation device estimates the position of the vehicle based on the map feature information and the calculated sensor feature information. As a result, the vehicle position estimation device can estimate the position of the vehicle with good accuracy based on the map feature information and the sensor feature information. In addition, since the map feature information is pre-stored in the map storage unit, the vehicle position estimation device can estimate the position of the vehicle at a higher speed.

[0019] Effects of the Invention

[0020] According to one aspect of the present disclosure, the position of the host vehicle can be estimated with high accuracy using reflected light of ambient light other than laser light in addition to reflected light of laser light. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a block diagram showing an example of a vehicle position estimating device according to an embodiment.

[0022] Figure 2 This is a flowchart showing the flow of the process of estimating the position of the host vehicle.

[0023] Figure 3This is a conceptual diagram showing the transition of various information used when estimating the position by the first method.

[0024] Figure 4 This is a conceptual diagram showing the transition of various information used when estimating the position by the second method.

[0025] Figure 5 This is a conceptual diagram showing the transition of various information used when estimating the position by the third method.

[0026] Figure 6 This is a conceptual diagram showing the transition of various information used when estimating the position by the fourth method.

[0027] Figure 7 This is a flowchart showing the flow of the estimation process for estimating the position of the host vehicle using the sixth method.

[0028] Description of symbols

[0029] 3 map storage unit, 11 laser irradiation unit, 12 light receiving element (light receiving unit), 14 light separation unit (light receiving unit), 15 laser processing unit (light receiving unit), 16 ambient light processing unit (light receiving unit), 21 own position estimating unit (estimating unit), 100 vehicle position estimating device. DETAILED DESCRIPTION

[0030] Hereinafter, exemplary embodiments will be described with reference to the accompanying drawings. In the drawings, the same or corresponding elements are denoted by the same reference numerals, and duplicate descriptions are omitted.

[0031] Figure 1 The vehicle position estimation device 100 shown is mounted on a vehicle (the vehicle itself) and estimates the position of the vehicle on a map. The information on the position of the vehicle estimated by the vehicle position estimation device 100 can be used for various controls such as automatic driving of the vehicle. The vehicle position estimation device 100 includes a laser radar [LIDAR: Light Detection and Ranging (light detection and ranging)] 1, an own position estimation ECU [Electronic Control Unit (electronic control unit)] 2, and a map storage unit 3.

[0032] The laser radar 1 irradiates the surroundings of the vehicle with a laser and receives the reflected light (the reflected light of the laser) after the irradiated laser is reflected by the target object. In addition, the laser radar 1 detects the intensity of the reflected light of the laser. The laser radar 1 in this embodiment can receive the reflected light of the ambient light (the reflected light of the ambient light) after the ambient light is reflected by the target object on the basis of the reflected light of the irradiated laser. The ambient light is light other than the irradiated laser. In addition, the laser radar 1 can detect the intensity of the reflected light of the received ambient light.

[0033] More specifically, the laser radar 1 includes a laser irradiation unit 11, a light receiving element 12, and a light processing ECU 13. The laser irradiation unit 11 irradiates laser light to a plurality of positions in a predetermined area around the vehicle on which the vehicle position estimating device 100 is mounted.

[0034] The light receiving element 12 can receive the reflected light of the laser irradiated from the laser irradiation unit 11, and output a signal corresponding to the intensity of the received reflected light of the laser. In addition, the light receiving element 12 can receive the reflected light of the ambient light other than the laser irradiated from the laser irradiation unit 11, and output a signal corresponding to the intensity of the received reflected light of the ambient light.

[0035] The light processing ECU 13 is an electronic control unit having a CPU, a ROM, a RAM, etc. The light processing ECU 13 realizes various functions by, for example, loading a program recorded in the ROM into the RAM and executing the program loaded into the RAM by the CPU. The light processing ECU 13 may also be composed of a plurality of electronic units.

[0036] The light processing ECU 13 detects the intensity of the reflected light of the laser light and the intensity of the reflected light of the ambient light respectively received by the light receiving element 12 based on the output signal of the light receiving element 12. The light processing ECU 13 functionally includes a light separation unit 14, a laser processing unit 15, and an ambient light processing unit 16. Thus, the light receiving element 12, the light separation unit 14, the laser processing unit 15, and the ambient light processing unit 16 function as a light receiving unit capable of receiving the reflected light of the laser light and the reflected light of the ambient light and detecting the intensity of each received reflected light.

[0037] The light separation unit 14 separates the light received by the light receiving element 12 into the reflected light of the laser and the reflected light of the ambient light. The light separation unit 14 can, for example, distinguish the light of a specific on-off pattern as the reflected light of the laser, and distinguish the light other than that as the reflected light of the ambient light. In addition, for example, the light separation unit 14 can distinguish the light received within a predetermined time from the laser irradiation unit 11 irradiating the laser as the reflected light of the laser, and distinguish the light received at a fixed time other than that as the reflected light of the ambient light. The predetermined time is pre-set based on the time from the laser irradiation unit 11 irradiating the laser to the time when the irradiated laser is reflected by the target object around the vehicle and the reflected light of the laser reaches the light receiving element 12. In addition, as described above, the reflected light of the ambient light does not include the reflected light of the laser irradiated from the laser radar 1. However, in the case where the ambient light includes light having a wavelength identical to that of the laser, the reflected light of the ambient light includes the reflected light of light having a wavelength identical to that of the laser.

[0038] The laser processing unit 15 generates laser information (a point cloud of the laser) based on the light reception result of the reflected light of the laser received by the light receiving element 12. The laser information is generated based on the light reception result of multiple lasers irradiated to each position in a predetermined irradiation area (the light reception result of multiple reflected lights). In addition, after the laser radar 1 completes the laser irradiation to all positions in the irradiation area, it irradiates the laser to each position in the irradiation area again. In this way, after the laser radar 1 completes the irradiation process of irradiating all positions in the irradiation area with the laser, it performs the next irradiation process again. Laser information is generated each time the laser radar 1 performs the irradiation process.

[0039] More specifically, the laser processing unit 15 generates laser point information by associating the three-dimensional position of the reflection point of the irradiated laser light with the intensity of the laser light for each of the plurality of laser light irradiated into the irradiation area. The laser processing unit 15 generates laser information based on the generated plurality of laser point information. In addition, the laser processing unit 15 can measure the three-dimensional position of the reflection point of the laser light based on the irradiation angle of the laser light irradiated from the laser irradiation unit 11 and the arrival time from the irradiation of the laser light to the arrival of the reflected light of the laser light at the light receiving element 12.

[0040] The ambient light processing unit 16 generates ambient light information (a point cloud of ambient light) based on the light reception result of the reflected light of the ambient light received by the light receiving element 12. The ambient light information is generated similarly to the laser information each time the laser radar 1 performs an irradiation process of irradiating a plurality of laser beams into the irradiation area.

[0041] In more detail, the ambient light processing unit 16 first obtains the three-dimensional position of the reflection point of the laser from the laser processing unit 15. Here, in a state where the irradiation angle of the laser and the states of the various parts of the laser radar 1 are not changed, the position of the reflection point of the laser received by the light receiving element 12 and the position of the reflection point of the ambient light become the same as each other. Therefore, the laser radar 1 detects the intensity of the reflected light of the ambient light when the reflected light of the laser is received, so that the intensity of the reflected light of the ambient light reflected at the same position as the reflection point of the laser can be detected. Therefore, the ambient light processing unit 16 generates ambient light point information by associating the three-dimensional position of the reflection point of the laser obtained from the laser processing unit 15 with the intensity of the reflected light of the ambient light received by the light receiving element 12. The ambient light point information is generated for each of the multiple lasers irradiated into the irradiation area. The ambient light processing unit 16 generates ambient light information based on the generated multiple ambient light point information.

[0042] Thus, the laser radar 1 can generate laser information and ambient light information based on the light reception result of the light receiving element 12. That is, the laser radar 1 can generate laser information and ambient light information based on the light reception result of one light receiving element 12, and therefore, calibration of the laser information and ambient light information is not required.

[0043] The map storage unit 3 stores in advance the position estimation information used when the self-position estimation ECU 2 estimates the position of the vehicle. The position estimation information is generated based on, for example, the laser information and ambient light information previously generated by the laser radar 1, and is stored in advance in the map storage unit 3. For example, the vehicle position estimation device 100 may measure the surroundings of the vehicle by the laser radar 1 while the vehicle is traveling, and store the position estimation information generated based on the laser information and ambient light information generated by the laser radar 1 in the map storage unit 3.

[0044] The own position estimation ECU 2 is an electronic control unit having the same structure as the light processing ECU 13. The own position estimation ECU 2 may be integrally formed with the light processing ECU 13 of the laser radar 1. The own position estimation ECU 2 functionally includes an own position estimation unit (estimation unit) 21.

[0045] The own position estimating unit 21 estimates the position of the own vehicle based on the light reception result of the reflected light of the laser (laser information) and the light reception result of the reflected light of the ambient light (ambient light information) received by the light receiving element 12. Here, the own position estimating unit 21 estimates the position of the own vehicle by comparing the light reception result of the reflected light of the laser and the ambient light with the position estimation information stored in the map storage unit 3. The details of the method for estimating the position of the own vehicle will be described later.

[0046] Next, use Figure 2 The flowchart of FIG. 1 illustrates the process of the estimation process of the position of the vehicle performed by the vehicle position estimation device 100. Figure 2 The process shown is repeatedly executed at predetermined intervals. Figure 2 As shown, the light separation unit 14 separates the light received by the light receiving element 12 into the reflected light of the laser and the reflected light of the ambient light (S101). The laser processing unit 15 calculates the three-dimensional position of the reflection point of the irradiated laser (S102). In addition, the laser processing unit 15 generates laser information, which is composed of a plurality of laser point information that associates the three-dimensional position of the reflection point of the irradiated laser with the intensity of the laser (S103).

[0047] The ambient light processing unit 16 obtains the three-dimensional position of the reflection point of the laser light calculated by the laser processing unit 15, and generates ambient light point information based on the obtained three-dimensional position of the reflection point and the intensity of the reflected light of the ambient light. In addition, the ambient light processing unit 16 generates ambient light information composed of the generated plurality of ambient light point information (S104). The own position estimating unit 21 estimates the position of the vehicle based on the generated laser information and ambient light information (S105).

[0048] Hereinafter, an example of various methods of estimating the position of the own vehicle performed by the own position estimating unit 21 will be described.

[0049] (Method 1)

[0050] First, the first method for estimating the position of the vehicle is described. Here, in the first method for estimating the position of the vehicle, the position estimation information stored in the map storage unit 3 is described. The map storage unit 3 pre-stores map laser information and map ambient light information as position estimation information for each location on the map.

[0051] The map laser information refers to information generated based on the light reception result of the reflected light of the laser. The map laser information stored in the map storage unit 3 and the laser information generated by the laser processing unit 15 are the same information that can be compared. Specifically, as an example, the laser information generated by the laser processing unit 15 in the past is stored in the map storage unit 3 as map laser information. In addition, the map ambient light information refers to information generated based on the light reception result of the reflected light of the ambient light. The map ambient light information stored in the map storage unit 3 and the ambient light information generated by the ambient light processing unit 16 are the same information that can be compared. Specifically, as an example, the ambient light information generated by the ambient light processing unit 16 in the past is stored in the map storage unit 3 as map ambient light information.

[0052] Below, use Figure 3The following is a conceptual diagram showing the transition of various information used when estimating the position by the first method. Figure 3 As shown, the self-position estimating unit 21 calculates laser matching information, which is the matching result of the laser information generated by the laser processing unit 15 of the laser radar 1 and the map laser information stored in the map storage unit 3. In addition, the self-position estimating unit 21 calculates ambient light matching information, which is the matching result of the ambient light information generated by the ambient light processing unit 16 of the laser radar 1 and the map ambient light information stored in the map storage unit 3.

[0053] Next, the own position estimation unit 21 calculates the own position score based on the calculated laser matching information and ambient light matching information. The own position score indicates the possibility (likelihood) that the own vehicle exists at a location on the map associated with the map laser information and the map ambient light information used when calculating the laser matching information and the ambient light matching information. When calculating the own position score, the own position estimation unit 21 uses the map laser information and the map ambient light information corresponding to a location (location on the map) different from the map laser information and the map ambient light information used when calculating the own position score this time to calculate the own position score. In this way, the own position estimation unit 21 calculates the own position score for each location on the map associated with the map laser information and the map ambient light information. And, based on the own position score of each location, the own position estimation unit 21 estimates the position of the own vehicle on the map by a well-known method.

[0054] (Specific example of the first method)

[0055] A more detailed specific example of the first method will be described.

[0056] (1) The self-position estimating unit 21 calculates the degree of consistency between the laser information generated by the laser processing unit 15 of the laser radar 1 and the map laser information stored in the map storage unit 3 as score 1. The self-position estimating unit 21 can calculate score 1 based on, for example, the degree of consistency between the three-dimensional position and reflection intensity value of the reflection point in the laser information and the map laser information. This score 1 is equivalent to the above-mentioned laser matching information.

[0057] (2) In addition, the self-position estimating unit 21 calculates the degree of consistency between the ambient light information generated by the ambient light processing unit 16 of the laser radar 1 and the map ambient light information stored in the map storage unit 3 as score 2. The self-position estimating unit 21 can calculate score 2 based on, for example, the degree of consistency between the three-dimensional position and reflection intensity value of the reflection point in the ambient light information and the map ambient light information. This score 2 is equivalent to the above-mentioned ambient light matching information.

[0058] (3) The own position estimation unit 21 calculates the comprehensive score based on the calculated score 1 and score 2. For example, the own position estimation unit 21 may calculate the comprehensive score as follows using the function f.

[0059] Comprehensive score = f(score 1, score 2)

[0060] In addition, the self-position estimation unit 21 calculates the self-position score based on the comprehensive score. Here, the self-position estimation unit 21 may also calculate the self-position score by, for example, multiplying the comprehensive score by a weight coefficient. The self-position estimation unit 21 calculates the self-position score for each location on the map associated with the map laser information and the map ambient light information.

[0061] (4) Furthermore, the own-vehicle position estimating unit 21 estimates the position of the own vehicle on the map by a well-known method based on the own-vehicle position score of each point.

[0062] (Method 2)

[0063] The second method for estimating the position of the vehicle is described. Here, in the second method for estimating the position of the vehicle, the position estimation information stored in the map storage unit 3 is described. The map storage unit 3, as in the case of the first method, stores map laser information and map ambient light information in advance for each location on the map as position estimation information.

[0064] Below, use Figure 4 The following is a conceptual diagram showing the transition of various information used when the position is estimated by the second method. Figure 4 As shown, the self-position estimating unit 21 integrates the laser information generated by the laser processing unit 15 of the laser radar 1 and the ambient light information generated by the ambient light processing unit 16, and calculates the integrated result as the sensor integrated information. In addition, the self-position estimating unit 21 integrates the map laser information and the map ambient light information stored in the map storage unit 3, and calculates the integrated result as the map integrated information.

[0065] Next, the own position estimation unit 21 calculates the own position score based on the calculated consistency between the sensor comprehensive information and the map comprehensive information. The own position score indicates the possibility (likelihood) that the own vehicle exists at a location on the map associated with the map laser information and the map ambient light information used when calculating the map comprehensive information. The own position estimation unit 21 calculates the own position score for each location on the map associated with the map laser information and the map ambient light information. And, the own position estimation unit 21 estimates the position of the own vehicle on the map by a well-known method based on the own position score of each location.

[0066] (Specific example of the second method)

[0067] A more detailed specific example of the second method will be described. First, a first example of the second method will be described.

[0068] (1) The self-position estimating unit 21 calculates an intensity score for each reflection point based on the intensity value of the reflected light of the laser in the laser information generated by the laser radar 1 and the intensity value of the reflected light of the ambient light in the ambient light information generated by the laser radar 1. The intensity score can also be calculated by performing a predetermined operation using the intensity value of the reflected light of the laser and the intensity value of the reflected light of the ambient light. In addition, the self-position estimating unit 21 generates sensor comprehensive information based on the three-dimensional position of the reflection point and the calculated intensity score. The sensor comprehensive information is composed of, for example, a data group of (X, Y, Z, intensity score).

[0069] (2) In addition, the self-position estimation unit 21 calculates a map intensity score for each reflection point based on the intensity value of the reflected light of the laser in the map laser information stored in the map storage unit 3 and the intensity value of the reflected light of the ambient light in the map ambient light information stored in the map storage unit 3. The intensity score can also be calculated by performing a predetermined operation using the intensity value of the reflected light of the laser and the intensity value of the reflected light of the ambient light. In addition, the self-position estimation unit 21 generates map comprehensive information based on the three-dimensional position of the reflection point and the calculated map intensity score. The map comprehensive information is composed of, for example, a data group of (X, Y, Z, map intensity score).

[0070] (3) The self-position estimation unit 21 calculates the self-position score based on the consistency between the generated sensor comprehensive information and the map comprehensive information. Here, the self-position estimation unit 21 can calculate the self-position score based on, for example, the consistency between the three-dimensional position of the reflection point and the consistency between the intensity score and the map intensity score. The self-position estimation unit 21 calculates the self-position score for each location on the map associated with the map laser information and the map ambient light information.

[0071] (4) Furthermore, the own-vehicle position estimating unit 21 estimates the position of the own vehicle on the map by a well-known method based on the own-vehicle position score of each point.

[0072] Next, the second example of the second method will be described.

[0073] (1) The self-position estimating unit 21 extracts feature quantities from laser information generated by the laser radar 1 and extracts feature quantities from ambient light information generated by the laser radar 1. The feature quantities are feature quantities used for object detection, for example, features that can be used to detect an object from an image.

[0074] (2) Next, the self-position estimating unit 21 assigns the attribute of the target object represented by the feature amount to the feature amount extracted from the laser information. In addition, the self-position estimating unit 21 assigns the attribute of the target object represented by the feature amount to the feature amount extracted from the ambient light information.

[0075] (3) The self-position estimation unit 21 integrates the laser information and the ambient light information to generate sensor comprehensive information. At that time, the self-position estimation unit 21 adds the attribute information of the feature quantity extracted from the laser information and the ambient light information to generate the sensor comprehensive information. When multiple feature quantities are extracted, the attribute information of multiple feature quantities is added. Furthermore, the self-position estimation unit 21 calculates the attribute score of the feature quantity and adds the calculated attribute score to the feature quantity to generate the sensor comprehensive information. The attribute score represents, for example, the likelihood of recognition of the attribute represented by the extracted feature quantity. For example, when the same feature quantity is extracted from the laser information and the ambient light information and the attributes are the same, the self-position estimation unit 21 calculates a high attribute score, and when the same feature quantity is extracted and the attributes are different, the self-position estimation unit 21 calculates a low attribute score. For example, when multiple feature quantities are extracted and the attributes of each feature quantity are "white line, sign, road surface, moving object", the self-position estimation unit 21 can also assign multi-class attribute scores to each feature quantity such as "white line: 0.7, sign: 0.05, road surface: 0.2, moving object: 0.1".

[0076] (4) The own position estimation unit 21 also extracts feature quantities and assigns attributes to the map laser information and map ambient light information stored in the map storage unit 3 in the same manner as in the above-mentioned (1) to (3), and further calculates attribute scores. In addition, the own position estimation unit 21 generates map comprehensive information obtained by integrating the map laser information and the map ambient light information in the same manner as the sensor comprehensive information to which the attributes and attribute scores are assigned.

[0077] (5) The self-position estimation unit 21 calculates the self-position score based on the calculated consistency between the sensor comprehensive information and the map comprehensive information. Here, the self-position estimation unit 21 can calculate the self-position score based on, for example, the consistency of the three-dimensional position of the reflection point and the consistency of the attribute score. The self-position estimation unit 21 calculates the self-position score for each location on the map associated with the map laser information and the map ambient light information.

[0078] (6) Furthermore, the own-vehicle position estimating unit 21 estimates the position of the own vehicle on the map by a well-known method based on the own-vehicle position score of each point.

[0079] In addition, in the second method, map laser information and map ambient light information are stored in advance in the map storage unit 3 for each location on the map as position estimation information. Not limited to this, the map storage unit 3 may also store map comprehensive information for each location on the map as position estimation information, and the map comprehensive information is a comprehensive result of the map laser information based on the light reception result of the reflected light of the laser and the map ambient light information based on the light reception result of the reflected light of the ambient light. That is, in the above-mentioned second method, the own position estimation unit 21 generates the map comprehensive information, but it is also possible that the map comprehensive information obtained by combining the map laser information and the map ambient light information is recorded in advance in the map storage unit 3.

[0080] In this case, the own position estimating unit 21 integrates the laser information generated by the laser processing unit 15 of the laser radar 1 and the ambient light information generated by the ambient light processing unit 16, and calculates the integrated result as the sensor integrated information. In addition, the own position estimating unit 21 can estimate the position of the own vehicle based on the generated sensor integrated information and the map integrated information stored in the map storage unit 3.

[0081] (Method 3)

[0082] The third method for estimating the position of the vehicle is described. Here, in the third method for estimating the position of the vehicle, the position estimation information stored in the map storage unit 3 is described. The map storage unit 3, as in the case of the first method, stores map laser information and map ambient light information in advance for each location on the map as position estimation information.

[0083] Below, use Figure 5 The following is a conceptual diagram showing the transition of various information used in position estimation by the third method. Figure 5 As shown, the self-position estimating unit 21 extracts the feature quantity for object detection from the laser information generated by the laser processing unit 15 of the laser radar 1 as the laser feature quantity, and extracts the feature quantity for object detection from the map laser information stored in the map storage unit 3 as the map laser feature quantity. The self-position estimating unit 21 calculates the matching result of the extracted laser feature quantity and the map laser feature quantity as the laser feature quantity information.

[0084] In addition, the self-position estimating unit 21 extracts a feature quantity for object detection from the ambient light information generated by the ambient light processing unit 16 of the laser radar 1 as the ambient light feature quantity, and extracts a feature quantity for object detection from the map ambient light information stored in the map storage unit 3 as the map ambient light feature quantity. The self-position estimating unit 21 calculates a matching result between the extracted ambient light feature quantity and the map ambient light feature quantity as the ambient light feature quantity information.

[0085] Furthermore, the own position estimation unit 21 calculates the own position score based on the consistency between the calculated laser characteristic quantity information and the ambient light characteristic quantity information. The own position score indicates the possibility (likelihood) that the own vehicle exists at a location on the map that associates the map laser information and the map ambient light information used when calculating the laser characteristic quantity information and the ambient light characteristic quantity information. The own position estimation unit 21 calculates the own position score for each location on the map that associates the map laser information and the map ambient light information. Furthermore, the own position estimation unit 21 estimates the position of the own vehicle on the map by a well-known method based on the own position score of each location.

[0086] (Specific example of the third method)

[0087] A more detailed specific example of the third method will be described.

[0088] (1) The self-position estimating unit 21 extracts laser feature quantities from the laser information generated by the laser processing unit 15 of the laser radar 1, and extracts map laser feature quantities from the map laser information stored in the map storage unit 3. The self-position estimating unit 21 calculates the consistency between the extracted laser feature quantity and the map laser feature quantity as score 1. The self-position estimating unit 21 can calculate score 1 based on, for example, the consistency of the point group of the feature quantity, the consistency of the shape of the point group of the feature quantity, and the consistency of the attribute of the feature quantity. This score 1 is equivalent to the above-mentioned laser feature quantity information.

[0089] (2) In addition, the self-position estimating unit 21 extracts the ambient light feature quantity from the ambient light information generated by the ambient light processing unit 16 of the laser radar 1, and extracts the map ambient light feature quantity from the map ambient light information stored in the map storage unit 3. The self-position estimating unit 21 calculates the consistency between the extracted ambient light feature quantity and the map ambient light feature quantity as score 2. The self-position estimating unit 21 can calculate score 2 based on, for example, the consistency of the point group of the feature quantity, the consistency of the shape of the point group of the feature quantity, and the consistency of the attribute of the feature quantity. This score 2 is equivalent to the above-mentioned ambient light feature quantity information.

[0090] (3) The own position estimation unit 21 calculates a feature quantity comprehensive score based on the calculated score 1 and score 2. For example, the own position estimation unit 21 may calculate the feature quantity comprehensive score as follows using the function f.

[0091] Feature quantity comprehensive score = f(score 1, score 2)

[0092] In addition, the self-position estimation unit 21 calculates the self-position score based on the feature quantity comprehensive score. Here, the self-position estimation unit 21 may also perform weighting such as multiplying the feature quantity comprehensive score by a weight coefficient to calculate the self-position score. The self-position estimation unit 21 calculates the self-position score for each location on the map associated with the map laser information and the map ambient light information.

[0093] (4) Furthermore, the own-vehicle position estimating unit 21 estimates the position of the own vehicle on the map by a well-known method based on the own-vehicle position score of each point.

[0094] Furthermore, in the third method, map laser information and map ambient light information are pre-stored in the map storage unit 3 as position estimation information for each location on the map. Not limited to this, map laser feature quantities extracted from the light reception result of the reflected light of the laser and map ambient light feature quantities extracted from the light reception result of the reflected light of the ambient light may also be pre-stored in the map storage unit 3 as position estimation information for each location on the map. That is, in the third method described above, the self-position estimation unit 21 extracts the map laser feature quantity and the map ambient light feature quantity, but it is also possible that the map laser feature quantity and the map ambient light feature quantity extracted from the map laser information and the map ambient light information, respectively, are pre-recorded in the map storage unit 3.

[0095] In this case, the own position estimation unit 21 extracts the laser feature quantity from the laser information generated by the laser processing unit 15 of the laser radar 1, and calculates the matching result between the map laser feature quantity stored in the map storage unit 3 and the extracted laser feature quantity as the laser feature quantity information. In addition, the own position estimation unit 21 extracts the ambient light feature quantity from the ambient light information generated by the ambient light processing unit 16 of the laser radar 1, and calculates the matching result between the map ambient light feature quantity stored in the map storage unit 3 and the extracted ambient light feature quantity as the ambient light feature quantity information. And, the own position estimation unit 21 can estimate the position of the own vehicle based on the calculated laser feature quantity information and the ambient light feature quantity information.

[0096] (Method 4)

[0097] The fourth method for estimating the position of the vehicle is described. Here, in the fourth method for estimating the position of the vehicle, the position estimation information stored in the map storage unit 3 is described. The map storage unit 3, as in the case of the first method, stores map laser information and map ambient light information in advance for each location on the map as position estimation information.

[0098] Below, use Figure 6 The following is a conceptual diagram showing the transition of various information used in position estimation by the fourth method. Figure 6As shown, the self-position estimating unit 21 extracts a feature quantity for detecting an object as a laser feature quantity from the laser information generated by the laser processing unit 15 of the laser radar 1, and extracts a feature quantity for detecting an object as an ambient light feature quantity from the ambient light information generated by the ambient light processing unit 16 of the laser radar 1. The self-position estimating unit 21 calculates a comprehensive result of the extracted laser feature quantity and ambient light feature quantity as sensor feature quantity information.

[0099] In addition, the self-position estimating unit 21 extracts a feature quantity for object detection from the map laser information stored in the map storage unit 3 as a map laser feature quantity, and extracts a feature quantity for object detection from the map ambient light information stored in the map storage unit 3 as a map ambient light feature quantity. The self-position estimating unit 21 calculates a comprehensive result of the extracted map laser feature quantity and map ambient light feature quantity as map feature quantity information.

[0100] Furthermore, the own position estimation unit 21 calculates the own position score based on the calculated consistency between the sensor feature quantity information and the map feature quantity information. The own position score indicates the possibility (likelihood) that the own vehicle exists at a location on the map associated with the map laser information and the map ambient light information used when calculating the map feature quantity information. The own position estimation unit 21 calculates the own position score for each location on the map associated with the map laser information and the map ambient light information. Furthermore, the own position estimation unit 21 estimates the position of the own vehicle on the map by a well-known method based on the own position score of each location.

[0101] (Specific example of the fourth method)

[0102] A more detailed example of the fourth method will be described. First, a first example of the fourth method will be described.

[0103] (1) The self-position estimating unit 21 extracts laser feature quantities from the laser information generated by the laser processing unit 15 of the laser radar 1, and extracts ambient light feature quantities from the ambient light information generated by the ambient light processing unit 16 of the laser radar 1. The self-position estimating unit 21 selects only mutually consistent feature quantities from the extracted laser feature quantities and ambient light feature quantities, and calculates the selected feature quantities as sensor feature quantity information.

[0104] (2) The self-position estimating unit 21 extracts the map laser feature quantity from the map laser information stored in the map storage unit 3, and extracts the map ambient light feature quantity from the map ambient light information stored in the map storage unit 3. The self-position estimating unit 21 selects only the mutually consistent feature quantities from the extracted map laser feature quantity and map ambient light feature quantity, and calculates the selected feature quantity as the map feature quantity information.

[0105] (3) The own position estimation unit 21 calculates the own position score by various well-known methods based on the calculated degree of agreement between the sensor feature information and the map feature information. The own position estimation unit 21 calculates the own position score for each point on the map associated with the map laser information and the map ambient light information.

[0106] (4) Furthermore, the own-vehicle position estimating unit 21 estimates the position of the own vehicle on the map by a well-known method based on the own-vehicle position score of each point.

[0107] Next, the second example of the fourth method will be described.

[0108] (1) The self-position estimating unit 21 extracts laser feature quantities from the laser information generated by the laser processing unit 15 of the laser radar 1, and extracts ambient light feature quantities from the ambient light information generated by the ambient light processing unit 16 of the laser radar 1. The self-position estimating unit 21 synthesizes the extracted laser feature quantities and ambient light feature quantities, and calculates the synthesized feature quantities as sensor feature quantity information. In addition, when calculating the sensor feature quantity information, the self-position estimating unit 21 assigns a score to the synthesized feature quantity to generate the sensor feature quantity information. For example, as the score assigned to the synthesized feature quantity, the self-position estimating unit 21 can increase the score for the feature quantity extracted from both the laser information and the ambient light information, and reduce the score for the feature quantity extracted only from the laser information and the ambient light information. For example, the self-position estimating unit 21 can also set the score to 3 for the feature quantity extracted from both the laser information and the ambient light information, set the score to 2 for the feature quantity extracted only from the laser information, and set the score to 1 for the feature quantity extracted only from the ambient light information.

[0109] (2) The own position estimation unit 21 also extracts the map laser feature quantity and the map ambient light feature quantity and calculates the score in the same manner as in the above-mentioned (1) with respect to the map laser information and the map ambient light information stored in the map storage unit 3. In addition, the own position estimation unit 21 calculates the map feature quantity information obtained by synthesizing the map laser feature quantity and the map ambient light feature quantity and assigning a score, similarly to the sensor feature quantity information obtained by assigning a score to the synthesized feature quantity.

[0110] (3) The own position estimation unit 21 calculates the own position score by various well-known methods based on the calculated degree of agreement between the sensor feature information and the map feature information. The own position estimation unit 21 calculates the own position score for each point on the map associated with the map laser information and the map ambient light information.

[0111] (4) Furthermore, the own-vehicle position estimating unit 21 estimates the position of the own vehicle on the map by a well-known method based on the own-vehicle position score of each point.

[0112] Furthermore, in the fourth method, map laser information and map ambient light information are pre-stored in the map storage unit 3 as position estimation information for each location on the map. Without limitation to this, map feature information may also be pre-stored in the map storage unit 3 as position estimation information for each location on the map, and the map feature information is a comprehensive result of the map laser feature extracted from the light reception result of the reflected light of the laser and the map ambient light feature extracted from the light reception result of the reflected light of the ambient light. That is, in the fourth method described above, the self-position estimation unit 21 calculates the map feature information, but the map feature information obtained by combining the map laser feature and the map ambient light feature may also be pre-recorded in the map storage unit 3.

[0113] In this case, the own position estimation unit 21 extracts the laser feature quantity from the laser information generated by the laser processing unit 15 of the laser radar 1, and extracts the ambient light feature quantity from the ambient light information generated by the ambient light processing unit 16 of the laser radar 1. The own position estimation unit 21 calculates the comprehensive result of the extracted laser feature quantity and the ambient light feature quantity as the sensor feature quantity information. In addition, the own position estimation unit 21 can estimate the position of the own vehicle based on the calculated sensor feature quantity information and the map feature quantity stored in the map storage unit 3.

[0114] (Method 5)

[0115] The fifth method for estimating the position of the vehicle is described. In the fifth method, the own position estimation unit 21 calculates the own position score based on the matching result of the ambient light feature extracted from the ambient light information generated by the laser radar 1 and the map ambient light feature extracted from the map ambient light information stored in the map storage unit 3. When calculating the own position score, the own position score is weighted based on the reflection intensity of the laser detected by the laser radar 1. The own position estimation unit 21 calculates the own position score for each location on the map associated with the map ambient light information. And, the own position estimation unit 21 estimates the position of the vehicle on the map by a well-known method based on the own position score of each location.

[0116] (Specific example of the fifth method)

[0117] A more detailed specific example of the fifth method will be described. First, a first example of the fifth method will be described.

[0118] (1) The self-position estimating unit 21 matches the ambient light feature quantity extracted from the ambient light information generated by the laser radar 1 with the map ambient light feature quantity extracted from the map ambient light information stored in the map storage unit 3. The self-position estimating unit 21 calculates the attribute of the consistent feature quantity as the result of the matching. When the attribute of the consistent feature quantity is correlated with the reflection intensity of the laser detected by the laser radar 1, the self-position estimating unit 21 calculates the self-position score by assigning a score that takes into account the reflection intensity of the laser to the feature quantity.

[0119] For example, when the attribute of the feature quantity is a white line or a sign, and the reflection intensity of the laser is high, a high score is assigned to the feature quantity, and the own position score becomes high. On the contrary, for example, when the attribute of the feature quantity is a white line or a sign, and the reflection intensity of the laser is low, a low score is assigned to the feature quantity, and the own position score becomes low. In addition, for example, when the attribute of the feature quantity is a road surface, a pole or a curb, and the reflection intensity of the laser is low, a high score is assigned to the feature quantity, and the own position score becomes high. On the contrary, for example, when the attribute of the feature quantity is a road surface, a pole or a curb, and the reflection intensity of the laser is high, a low score is assigned to the feature quantity, and the own position score becomes low.

[0120] (2) The own position estimation unit 21 calculates the own position score for each point on the map associated with the map ambient light information.

[0121] (3) Furthermore, the own-vehicle position estimating unit 21 estimates the position of the own vehicle on the map by a well-known method based on the own-vehicle position score of each point.

[0122] Next, the second example of the fifth method will be described.

[0123] (1) The self-position estimating unit 21 matches the ambient light feature extracted from the ambient light information generated by the laser radar 1 with the map ambient light feature extracted from the map ambient light information stored in the map storage unit 3. For each point of the consistent feature (each point of the reflected light), the self-position estimating unit 21 calculates a self-position score obtained by assigning a high score to the feature when the distribution of the reflection intensity value of the laser detected by the laser radar 1 is within a certain range. On the contrary, for each point of the consistent feature, the self-position estimating unit 21 calculates a self-position score obtained by assigning a low score to the feature when the distribution of the reflection intensity value of the laser detected by the laser radar 1 is not within a certain range. In this way, the self-position score changes according to the score assigned to the feature.

[0124] (2) The own position estimation unit 21 calculates the own position score for each point on the map associated with the map ambient light information.

[0125] (3) Furthermore, the own-vehicle position estimating unit 21 estimates the position of the own vehicle on the map by a well-known method based on the own-vehicle position score of each point.

[0126] In addition, in the first and second examples described above, the case where the feature quantity about the ambient light is extracted and the self-position score is weighted based on the reflection intensity of the laser is described as an example. In contrast, the self-position estimating unit 21 calculates the self-position score based on the matching result of the laser feature quantity extracted from the laser information generated by the laser radar 1 and the map laser feature quantity extracted from the map laser information stored in the map storage unit 3. In addition, when calculating the self-position score, the self-position score may be weighted based on the reflection intensity of the ambient light detected by the laser radar 1.

[0127] (Method 6)

[0128] The sixth method for estimating the position of the own vehicle is described. In the sixth method, when the own position estimating unit 21 calculates the own position score using the feature value related to the laser and estimates the position of the own vehicle, if the calculated own position score is low, the position of the own vehicle is estimated using the reflected light of the ambient light.

[0129] (Specific example of the sixth method)

[0130] use Figure 7 The flowchart of is used to illustrate an example of a more detailed specific example in the sixth method. In addition, Figure 7 In the illustrated process, when the process of estimating the position of the host vehicle is started and the process reaches "end", the process is restarted from "start" after a predetermined time.

[0131] (1) Figure 7 As shown, the own position estimating unit 21 divides the area when estimating the position of the own vehicle based on the position estimation information of the map storage unit 3 into multiple search areas. The own position estimating unit 21 selects representative points in the search area for each of the divided multiple search areas (S201). The own position estimating unit 21 uses the feature quantity about the laser to calculate the own position score of each representative point (S202). Specifically, for example, the own position estimating unit 21 calculates the own position score of each representative point based on the laser feature quantity extracted from the laser information generated by the laser processing unit 15 of the laser radar 1 and the map laser feature quantity of each representative point extracted from the map laser information stored in the map storage unit 3 by various well-known methods.

[0132] The self-position estimating unit 21 determines whether the calculated self-position score of each representative point is greater than a predetermined score threshold (S203). Here, for example, the self-position estimating unit 21 may determine whether the average of the self-position scores of each representative point is greater than the score threshold.

[0133] When the calculated own position score is greater than the score threshold value (S203: Yes), the own position estimating unit 21 estimates the position of the own vehicle by various well-known methods based on the laser feature extracted from the laser information and the map laser feature of each point on the map extracted from the map laser information (S204). Here, the own position estimating unit 21 estimates the position of the own vehicle based on the map laser feature of each point on the map extracted from the map laser information, not the feature of the representative point.

[0134] On the other hand, when the calculated own position score is below the score threshold (S203: No), the own position estimation unit 21 uses the feature quantity about the ambient light to calculate the own position score of each representative point according to each feature quantity (S205). Specifically, for example, the own position estimation unit 21 calculates the own position score of each representative point by various well-known methods based on the ambient light feature quantity extracted from the ambient light information generated by the ambient light processing unit 16 of the laser radar 1 and the map ambient light feature quantity of each representative point extracted from the map ambient light information stored in the map storage unit 3. In addition, when calculating the own position score of each representative point, the own position estimation unit 21 calculates the own position score according to each extracted feature quantity (each ambient light feature quantity and the map ambient light feature quantity).

[0135] The self-position estimating unit 21 selects a feature quantity for estimating the position of the vehicle together with the feature quantity about the laser from the feature quantity about the ambient light used when calculating the self-position score (S206). For example, the self-position estimating unit 21 may select a predetermined number of feature quantities in sequence starting from the feature quantity with the highest self-position score. In addition, the self-position estimating unit 21 may select a predetermined number of feature quantities. Alternatively, the self-position estimating unit 21 may select the feature quantity about the ambient light in the order of the self-position score from high to low when adding the feature quantity about the ambient light to the feature quantity about the laser to estimate the self-position, until the self-position score exceeds the predetermined reference score threshold.

[0136] The own position estimating unit 21 estimates the position of the own vehicle by various well-known methods based on the selected feature quantities of ambient light (ambient light feature quantity and map ambient light feature quantity) on the basis of the laser feature quantity extracted from the laser information and the map laser feature quantity of each point on the map extracted from the map laser information (S207). Here, the own position estimating unit 21 estimates the position of the own vehicle based on the feature quantities of each point on the map extracted from the map laser information and the map ambient light information, not the feature quantity of the representative point.

[0137] In addition, in the above description, the case where the self-position score is calculated using the feature quantity related to the laser, and the reflected light of the ambient light is used when the calculated self-position score is low is described as an example. On the contrary, the self-position estimating unit 21 may calculate the self-position score using the feature quantity related to the ambient light, and estimate the position of the host vehicle using the reflected light of the laser when the calculated self-position score is low.

[0138] As described above, the vehicle position estimation device 100 includes the own position estimation unit 21, which estimates the position of the own vehicle based on the light reception result of the reflected light of the laser and the light reception result of the reflected light of the ambient light. For example, even when there is an object target that is difficult to detect with a laser, it is sometimes possible to detect the object target based on the reflected light of the ambient light. Therefore, the vehicle position estimation device 100 can estimate the position of the own vehicle with good accuracy by using the reflected light of the ambient light other than the laser in addition to the reflected light of the laser.

[0139] When estimating the position of the own vehicle by the above-mentioned first method, the own position estimating unit 21 calculates laser matching information, which is the matching result of the laser information generated by the laser processing unit 15 and the map laser information stored in the map storage unit 3, and calculates ambient light matching information, which is the matching result of the ambient light information generated by the ambient light processing unit 16 and the map ambient light information stored in the map storage unit 3. And the own position estimating unit 21 estimates the position of the own vehicle based on the calculated laser matching information and ambient light matching information. In this case, the vehicle position estimating device 100 can estimate the position of the own vehicle with good accuracy using the laser matching information and the ambient light matching information.

[0140] When estimating the position of the host vehicle by the second method described above, the own position estimating unit 21 integrates the laser information and the ambient light information generated by the laser radar 1 to calculate the sensor comprehensive information, and integrates the map laser information and the map ambient light information stored in the map storage unit 3 to calculate the map comprehensive information. Furthermore, the own position estimating unit 21 estimates the position of the host vehicle based on the calculated sensor comprehensive information and map comprehensive information. In this case, the vehicle position estimating device 100 can estimate the position of the host vehicle with good accuracy based on the sensor comprehensive information and the map comprehensive information.

[0141] Here, when estimating the position of the host vehicle by the second method, the map comprehensive information obtained by integrating the map laser information and the map ambient light information may be recorded in advance in the map storage unit 3 for each location on the map. Even in this case, the vehicle position estimation device 100 can estimate the position of the host vehicle with good accuracy based on the sensor comprehensive information generated based on the light receiving result of the laser radar 1 and the map comprehensive information stored in the map storage unit 3. In addition, since the map comprehensive information is pre-stored in the map storage unit 3, the vehicle position estimation device 100 can estimate the position of the host vehicle at a higher speed.

[0142] When estimating the position of the vehicle by the third method described above, the own position estimating unit 21 estimates the position of the vehicle by using the feature quantity used for object detection. Here, the own position estimating unit 21 matches the laser feature quantity extracted from the laser information generated by the laser radar 1 and the map laser feature quantity extracted from the map laser information stored in the map storage unit 3 to calculate the laser feature quantity information. In addition, the own position estimating unit 21 matches the ambient light feature quantity extracted from the ambient light information generated by the laser radar 1 and the map ambient light feature quantity extracted from the map ambient light information stored in the map storage unit 3 to calculate the ambient light feature quantity information. And, the own position estimating unit 21 estimates the position of the vehicle based on the calculated laser feature quantity information and the ambient light feature quantity information. In this case, the vehicle position estimating device 100 can estimate the position of the vehicle with good accuracy based on the laser feature quantity information and the ambient light feature quantity information.

[0143] Here, it is also possible that, when estimating the position of the own vehicle by the third method, the map laser feature quantity extracted from the map laser information and the map ambient light feature quantity extracted from the map ambient light information are pre-stored in the map storage unit 3 according to each location on the map. In this case, the own position estimation unit 21 matches the laser feature quantity extracted from the laser information generated by the laser radar 1 with the map laser feature quantity stored in the map storage unit 3 to calculate the laser feature quantity information. In addition, the own position estimation unit 21 matches the ambient light feature quantity extracted from the ambient light information generated by the laser radar 1 with the map laser feature quantity stored in the map storage unit 3 to calculate the laser feature quantity information. And, the own position estimation unit 21 estimates the position of the own vehicle based on the calculated laser feature quantity information and ambient light feature quantity information. Even in this case, the vehicle position estimation device can estimate the position of the own vehicle with good accuracy based on the laser feature quantity information and the ambient light feature quantity information. Furthermore, since the map laser feature quantity and the map ambient light feature quantity are stored in advance in the map storage unit 3 , the vehicle position estimating device 100 can estimate the position of the host vehicle at a higher speed.

[0144] When estimating the position of the vehicle by the fourth method described above, the own position estimating unit 21 estimates the position of the vehicle by using the feature quantity used for object detection. Here, the own position estimating unit 21 calculates the sensor feature quantity information by synthesizing the laser feature quantity extracted from the laser information generated by the laser radar 1 and the ambient light feature quantity extracted from the ambient light information. In addition, the own position estimating unit 21 calculates the map feature quantity information by synthesizing the map laser feature quantity extracted from the map laser information stored in the map storage unit 3 and the map ambient light feature quantity extracted from the map ambient light information. And, the own position estimating unit 21 estimates the position of the vehicle based on the calculated sensor feature quantity information and map feature quantity information. In this case, the vehicle position estimating device 100 can estimate the position of the vehicle with good accuracy based on the sensor feature quantity information and the map feature quantity information.

[0145] Here, when estimating the position of the own vehicle by the fourth method, the map feature quantity information obtained by integrating the map laser feature quantity extracted from the map laser information and the map ambient light feature quantity extracted from the map ambient light information is pre-stored in the map storage unit 3 for each location on the map. In this case, the own position estimation unit 21 calculates the sensor feature quantity information by integrating the laser feature quantity and the ambient light feature quantity respectively extracted from the laser information and the ambient light information generated by the laser radar 1. And, the own position estimation unit 21 estimates the position of the own vehicle based on the calculated sensor feature quantity information and the map feature quantity information stored in the map storage unit 3. Even in this case, the vehicle position estimation device 100 can estimate the position of the own vehicle with good accuracy based on the sensor feature quantity information and the map feature quantity information. In addition, since the map feature quantity information is pre-stored in the map storage unit 3, the vehicle position estimation device 100 can estimate the position of the own vehicle at a higher speed.

[0146] Although the embodiments of the present disclosure have been described above, the present disclosure is not limited to the above embodiments. The present disclosure can be modified in various ways without departing from the spirit of the present disclosure.

Claims

1. A vehicle position estimation device for estimating the position of a vehicle, comprising: a laser irradiation unit that irradiates laser light; a light receiving unit capable of receiving the reflected light of the laser light and the reflected light of ambient light which is light other than the laser light, and detecting the intensity of the received reflected light of the laser light and the intensity of the reflected light of the ambient light; an estimating unit that estimates the position of the host vehicle based on a result of receiving the reflected light of the laser light and a result of receiving the reflected light of the ambient light received by the light receiving unit; as well as a map storage unit, wherein the map storage unit pre-stores map laser information based on the light reception result of the reflected light of the laser and map ambient light information based on the light reception result of the reflected light of the ambient light for each location on the map, The estimating unit is configured as follows: dividing an area for estimating the position of the vehicle based on the position estimation information for estimating the position of the vehicle stored in the map storage unit into a plurality of search areas, selecting a representative point in each of the plurality of divided search areas, extracting a laser feature quantity from laser information based on a light reception result of the reflected light of the laser received by the light receiving unit, extracting a map laser feature quantity of each representative location from the map laser information stored in the map storage unit, and calculating a self-position score of each representative location using the laser feature quantity and the map laser feature quantity, When the calculated own position score is greater than the score threshold, estimating the position of the vehicle based on the laser characteristic amount and the map laser characteristic amount, When the calculated own position score is below the score threshold, extracting ambient light feature quantities from ambient light information based on the light reception result of the reflected light of the ambient light received by the light receiving unit, extracting map ambient light feature quantities of each representative location from the map ambient light information stored in the map storage unit, and calculating the own position score of each representative point according to the extracted ambient light feature quantities and the map ambient light feature quantities, From the ambient light feature quantities and the map ambient light feature quantities used in calculating the own position score, a predetermined number of the ambient light feature quantities and the map ambient light feature quantities are selected in order from the feature quantity with the highest own position score, Based on the laser feature extracted from the laser information and the map laser feature of each point on the map extracted from the map laser information, the position of the vehicle is estimated based on the selected ambient light feature and the map ambient light feature.

Images