JP2019160003A5 - Obstacle recognition system, server, obstacle recognition device, and database construction method - Google Patents
Obstacle recognition system, server, obstacle recognition device, and database construction method Download PDFInfo
- Publication number
- JP2019160003A5 JP2019160003A5 JP2018047742A JP2018047742A JP2019160003A5 JP 2019160003 A5 JP2019160003 A5 JP 2019160003A5 JP 2018047742 A JP2018047742 A JP 2018047742A JP 2018047742 A JP2018047742 A JP 2018047742A JP 2019160003 A5 JP2019160003 A5 JP 2019160003A5
- Authority
- JP
- Japan
- Prior art keywords
- obstacle
- likelihood
- recognition device
- unit
- result
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000010276 construction Methods 0.000 title 1
- 238000001514 detection method Methods 0.000 claims description 19
- 230000006399 behavior Effects 0.000 claims 5
- 230000000875 corresponding Effects 0.000 claims 4
- 238000003384 imaging method Methods 0.000 claims 1
Description
本開示の一形態によれば、移動可能な移動装置（１００）に搭載されサーバ（２５０）と通信可能な障害物認識装置（６０）が提供される。この障害物認識装置は、前記移動装置の進行方向における物標の検出結果を取得して、前記進行方向に前記移動装置との衝突回避が必要な障害物が存在する尤度を、予め定められた尤度算出ロジックを用いて算出する尤度算出部（４０）と、前記尤度が第１閾値（Ｘ）以下である場合に前記障害物なしと判断し、前記尤度が前記第１閾値よりも大きい第２閾値（Ｙ）以上である場合に前記障害物ありと判断し、前記尤度が前記第１閾値よりも大きく前記第２閾値よりも小さい場合に、前記障害物ありと暫定的に判断することである暫定的障害物ありと判断する、判断部（５０）と、を備える。前記判断部が前記暫定的障害物ありと判断した場合に、前記サーバに、前記検出結果と前記尤度算出部の算出結果とを送信し、前記暫定的障害物ありという判断が前記障害物なし又は前記障害物ありに修正された修正結果を前記サーバから受信してよい。 According to an aspect of the present disclosure, there is provided an obstacle recognition device (60) mounted on a movable mobile device (100) and capable of communicating with a server (250). The obstacle recognition device obtains a detection result of a target in the traveling direction of the moving device, and determines in advance the likelihood that there is an obstacle that needs to avoid a collision with the moving device in the traveling direction. A likelihood calculation unit (40) that calculates using a likelihood calculation logic, and if the likelihood is equal to or less than a first threshold value (X), it is determined that there is no obstacle, and the likelihood is the first threshold value. If the likelihood is greater than or equal to a second threshold value (Y) that is greater than the threshold value, and the likelihood is greater than the first threshold value and less than the second threshold value, the obstacle is provisionally present. And a determination unit (50) that determines that there is a temporary obstacle. When the determination unit determines that the temporary obstacle is present, the detection result and the calculation result of the likelihood calculation unit are transmitted to the server, and the determination that the temporary obstacle is present does not include the obstacle. Alternatively, the correction result corrected with the obstacle may be received from the server.
Claims (12)
前記移動装置の進行方向における物標の検出結果を取得して、前記進行方向に前記移動装置との衝突回避が必要な障害物が存在する尤度を、予め定められた尤度算出ロジックを用いて算出する尤度算出部（４０）と、
前記尤度が第１閾値（Ｘ）以下である場合に前記障害物なしと判断し、前記尤度が前記第１閾値よりも大きい第２閾値（Ｙ）以上である場合に前記障害物ありと判断し、前記尤度が前記第１閾値よりも大きく前記第２閾値よりも小さい場合に、前記障害物ありと暫定的に判断することである暫定的障害物ありと判断する、判断部（５０）と、を備え、
前記判断部が前記暫定的障害物ありと判断した場合に、前記サーバに、前記検出結果と前記尤度算出部の算出結果とを送信し、
前記暫定的障害物ありという判断が前記障害物なし又は前記障害物ありに修正された修正結果を前記サーバから受信する、障害物認識装置。 An obstacle recognition device (60 ) mounted on a movable mobile device (100) and capable of communicating with a server (250) ,
Acquires the detection result of the target object in the traveling direction before Symbol mobile device, the likelihood of collision avoidance obstacle required with the mobile device to the traveling direction is present, a predetermined likelihood calculation logic A likelihood calculation unit (40) for calculation using
When the likelihood is less than or equal to a first threshold value (X), it is determined that there is no obstacle, and when the likelihood is greater than or equal to a second threshold value (Y) that is greater than the first threshold value, it means that there is an obstacle. determining, when the likelihood is smaller than the larger second threshold value than the first threshold value, it is determined that there is pre-Symbol provisional obstacle is to provisionally determined that there is an obstacle, the determining unit ( 50) and
When the determination unit determines that the provisional obstacle is present, the detection result and the calculation result of the likelihood calculation unit are transmitted to the server,
An obstacle recognition device that receives from the server a correction result in which the determination that the temporary obstacle is present is corrected to the absence of the obstacle or the presence of the obstacle.
前記検出結果と前記算出結果とを用いて再構築された尤度算出ロジックを前記サーバから受信し、前記尤度算出部が用いる尤度算出ロジックを前記サーバから受信した尤度算出ロジックに更新する、障害物認識装置。 The likelihood calculation logic reconstructed using the detection result and the calculation result is received from the server, and the likelihood calculation logic used by the likelihood calculation unit is updated to the likelihood calculation logic received from the server. , Obstacle recognition device.
前記更新後の尤度算出ロジックは、前記更新前の尤度算出ロジックよりも、前記尤度が、前記第１閾値よりも大きく前記第２閾値よりも小さい値になりにくい、障害物認識装置。 The updated likelihood calculation logic is an obstacle recognition device in which the likelihood is less likely to be a value larger than the first threshold and smaller than the second threshold than the likelihood calculation logic before the update.
前記判断部による判断結果に基づいて運転継続の可否を判断する制御部（２０）を備え、 A control unit (20) for determining whether or not the operation can be continued based on the determination result by the determination unit,
前記判断部が前記暫定的障害物ありと判断し、かつ、前記制御部が運転継続を不可能であると判断した場合に、前記修正結果を受信するまで、前記移動装置を停止させるか前記移動装置の自動運転機能をオフにする、障害物認識装置。 When the determination unit determines that the temporary obstacle is present and when the control unit determines that the operation cannot be continued, the mobile device is stopped or the movement is continued until the correction result is received. An obstacle recognition device that turns off the automatic driving function of the device.
前記検出結果は、前記進行方向における物標を検出する複数のセンサ（３１、３２、３３）の検出結果であり、
前記尤度算出部は、前記複数のセンサごとに算出した前記尤度を合成し、
前記判断部は、合成された前記尤度を用いて前記判断を行う、障害物認識装置。 The obstacle recognition device according to any one of claims 1 to 4 ,
The detection result is a detection result of a plurality of sensors (31, 32, 33) for detecting a target in the traveling direction,
The likelihood calculation unit synthesizes the likelihood calculated for each of the plurality of sensors,
The obstacle recognition device , wherein the determination unit makes the determination using the combined likelihood .
前記尤度算出部は、前記進行方向に対する垂直断面のうち、物標が検出された垂直断面において、前記検出結果を複数の領域に分割して、分割された領域ごとに前記尤度を算出し、
前記判断部は、前記領域ごとに前記判断を行い、
前記複数の領域のうち、前記暫定的障害物ありと判断された領域の前記尤度を前記算出結果として前記サーバに送信する、障害物認識装置。 The obstacle recognition device according to any one of claims 1 to 5 ,
The likelihood calculation unit divides the detection result into a plurality of regions in a vertical cross section in which a target is detected among vertical cross sections with respect to the traveling direction, and calculates the likelihood for each of the divided regions. ,
The determination unit performs the determination for each of the areas,
Among previous SL plurality of regions, transmitting the likelihood of realm it is determined that there is the preliminary obstacle to the server as the calculation result, the obstacle recognition device.
前記移動装置の進行方向における物標の検出結果に基づいて算出された、前記進行方向に前記移動装置との衝突回避が必要な障害物が存在する尤度が、前記障害物なしを示す第１閾値以下でも、前記障害物ありを示す前記第１閾値より大きな第２閾値以上でもなく、前記障害物ありと暫定的に判断することである暫定的障害物ありを示す、前記第１閾値より大きく前記第２閾値より小さな値になった場合に、前記検出結果と前記尤度の算出結果とを前記障害物認識装置から受信し、物標と前記障害物との対応関係と、前記検出結果における物標と、の類似度を用いて、前記暫定的障害物ありを、前記障害物なし又は前記障害物ありに修正する修正部（２２０）、を備え、 The likelihood that an obstacle that needs to avoid a collision with the moving device in the moving direction is calculated based on the detection result of the target in the moving direction of the moving device indicates that there is no obstacle. It is less than or equal to a threshold value and is greater than or equal to a second threshold value that is greater than the first threshold value that indicates the presence of the obstacle, and is provisionally determined to be the presence of the obstacle. When the value is smaller than the second threshold value, the detection result and the likelihood calculation result are received from the obstacle recognition device, and the correspondence between the target and the obstacle and the detection result A correction unit (220) for correcting the presence of the temporary obstacle to the absence of the obstacle or the presence of the obstacle by using the similarity with the target,
前記修正部による修正結果を前記障害物認識装置に送信する、サーバ。 A server that transmits the correction result by the correction unit to the obstacle recognition device.
前記障害物認定装置から受信した、前記暫定的障害物ありに対応する、前記検出結果および前記算出結果と、前記修正部による前記暫定的障害物ありの修正結果と、を教師あり学習のためのデータとして格納する格納部（２４０）と、 For the supervised learning, the detection result and the calculation result, which are received from the obstacle recognizing device and correspond to the provisional obstacle, and the correction result of the provisional obstacle by the correction unit, are provided. A storage unit (240) for storing as data,
前記格納部に格納された前記教師あり学習のためのデータを用いて教師あり学習を行い、尤度算出ロジックを再構築する学習部（２３０）と、 A learning unit (230) for performing supervised learning using the data for supervised learning stored in the storage unit and reconstructing a likelihood calculation logic;
前記尤度算出ロジックを、前記障害物認識装置に送信して更新させる更新部（２４５）と、を備える、サーバ。 An updating unit (245) for transmitting the likelihood calculation logic to the obstacle recognition device to update it, the server.
前記学習部は、前記暫定的障害物ありが前記障害物なしに修正された場合に、前記暫定的障害物ありに対応する前記尤度算出部の算出結果が前記第１閾値以下となり、前記暫定的障害物ありが前記障害物ありに修正された場合に、前記暫定的障害物ありに対応する前記尤度算出部の算出結果が前記第２閾値以上となるように、前記尤度算出ロジックを再構築する、サーバ。 The learning unit, when the provisional obstacle is corrected without the obstacle, the calculation result of the likelihood calculation unit corresponding to the provisional obstacle is less than or equal to the first threshold value, and the provisional obstacle is obtained. When the presence of the obstacle is corrected to the presence of the obstacle, the likelihood calculation logic is set so that the calculation result of the likelihood calculation unit corresponding to the provisional obstacle is equal to or more than the second threshold value. The server to rebuild.
前記対応関係は、さらに、予め定められた期間経過後の前記障害物の挙動が予め定められた第１挙動であるか否かを含み、 The correspondence further includes whether or not the behavior of the obstacle after a predetermined period of time is a predetermined first behavior,
前記修正部は、さらに、前記暫定的障害物ありを前記障害物ありに修正した場合であって、かつ、前記暫定的障害物ありに対応する前記障害物の挙動が前記第１挙動である場合に、前記第１挙動を前記障害物の属性として付与し、 The correction unit further corrects the provisional obstacle existence to the existence of the obstacle, and the behavior of the obstacle corresponding to the provisional obstacle existence is the first behavior. To the first behavior as an attribute of the obstacle,
前記格納部は、さらに、前記属性が付与された前記教師あり学習のためのデータを格納し、 The storage section further stores data for the supervised learning to which the attribute is added,
前記学習部は、前記属性が付与された前記教師あり学習のためのデータに重み付けをして前記教師あり学習を行う、サーバ。 The learning unit weights data for the supervised learning to which the attribute is added, and performs the supervised learning.
移動装置の進行方向における物標の検出結果を用いて、前記進行方向に前記移動装置との衝突回避が必要な障害物が存在する尤度を、予め定められた尤度算出ロジックを用いて算出し、
前記尤度が第１閾値以下である場合に前記障害物なしと判断し、前記尤度が前記第１閾値よりも大きい第２閾値以上である場合に前記障害物あり判断し、前記尤度が前記第１閾値よりも大きく前記第２閾値よりも小さい場合に、前記障害物ありと暫定的に判断することである暫定的障害物ありと判断し、
前記暫定的障害物ありに対応する、前記進行方向の撮像結果と前記尤度の算出結果とを取得し、物標と前記障害物との予め定められた対応関係と、前記検出結果における物標と、の類似度を用いて、前記暫定的障害物ありを、前記障害物なし又は前記障害物ありに修正し、
前記暫定的障害物ありに対応する、前記検出結果と前記尤度の算出結果と、前記暫定的障害物ありの修正結果と、を教師あり学習のためのデータとして格納する、方法。 A method of constructing a database for storing supervised data for constructing a likelihood calculation logic, comprising:
Using the detection result of the target in the traveling direction of the moving device, the likelihood that an obstacle requiring collision avoidance with the moving device exists in the traveling direction is calculated using a predetermined likelihood calculation logic. Then
When the likelihood is less than or equal to a first threshold, it is determined that there is no obstacle, when the likelihood is greater than or equal to a second threshold that is greater than the first threshold, it is determined that there is an obstacle, and the likelihood is If the less than greater the second threshold value than the first threshold value, determines that there is pre-Symbol provisional obstacle is to provisionally determined that there is an obstacle,
Acquiring the imaging result in the traveling direction and the calculation result of the likelihood, which correspond to the provisional obstacle, and a predetermined correspondence between the target and the obstacle, and the target in the detection result. By using the similarity of, the provisional obstacle is corrected to the absence of the obstacle or the presence of the obstacle,
A method of storing the detection result, the calculation result of the likelihood, and the correction result of the temporary obstacle corresponding to the existence of the temporary obstacle as data for supervised learning.
前記障害物認識装置と通信可能なサーバ（２５０）と、を備える障害物認識システム（１０）であって、 An obstacle recognition system (10) comprising a server (250) capable of communicating with the obstacle recognition device,
前記障害物認識装置は、 The obstacle recognition device,
前記移動装置の進行方向における物標の検出結果を取得して、前記進行方向に前記移動装置との衝突回避が必要な障害物が存在する尤度を、予め定められた尤度算出ロジックを用いて算出する尤度算出部（４０）と、 The detection result of the target in the traveling direction of the moving device is acquired, and the likelihood that an obstacle that needs to avoid a collision with the moving device exists in the traveling direction is calculated using a predetermined likelihood calculation logic. A likelihood calculating unit (40) for calculating
前記尤度が第１閾値（Ｘ）以下である場合に前記障害物なしと判断し、前記尤度が前記第１閾値よりも大きい第２閾値（Ｙ）以上である場合に前記障害物ありと判断し、前記尤度が前記第１閾値よりも大きく前記第２閾値よりも小さい場合に、前記障害物有と暫定的に判断することである暫定的障害物ありと判断する、判断部（５０）と、を備え、 When the likelihood is less than or equal to a first threshold value (X), it is determined that there is no obstacle, and when the likelihood is greater than or equal to a second threshold value (Y) that is greater than the first threshold value, it means that there is an obstacle. A determination unit (50) that determines that there is a temporary obstacle, which is to tentatively determine that the obstacle exists if the likelihood is larger than the first threshold and smaller than the second threshold. ) And,
前記判断部が前記暫定的障害物ありと判断した場合に、前記サーバに、前記検出結果と前記尤度算出部の算出結果とを送信し、 When the determination unit determines that the temporary obstacle is present, the detection result and the calculation result of the likelihood calculation unit are transmitted to the server,
前記サーバは、 The server is
前記障害物認識装置から受信した前記検出結果および前記算出結果と、物標と前記障害物との対応関係と、前記検出結果における物標と、の類似度を用いて、前記暫定的障害物ありを、前記障害物なし又は前記障害物ありに修正する修正部（２２０）を備え、 Using the similarity between the detection result and the calculation result received from the obstacle recognition device, the correspondence between the target and the obstacle, and the target in the detection result, the provisional obstacle is present. Is provided with a correction unit (220) that corrects
前記修正部による修正結果を前記障害物認識装置に送信する、 Transmitting the correction result by the correction unit to the obstacle recognition device,
障害物認識システム（１０）。Obstacle recognition system (10).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2018047742A JP6996368B2 (en) | 2018-03-15 | 2018-03-15 | Server, obstacle recognition device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2018047742A JP6996368B2 (en) | 2018-03-15 | 2018-03-15 | Server, obstacle recognition device |
Publications (3)
| Publication Number | Publication Date |
|---|---|
| JP2019160003A JP2019160003A (en) | 2019-09-19 |
| JP2019160003A5 true JP2019160003A5 (en) | 2020-07-27 |
| JP6996368B2 JP6996368B2 (en) | 2022-01-17 |
Family
ID=67994950
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP2018047742A Active JP6996368B2 (en) | 2018-03-15 | 2018-03-15 | Server, obstacle recognition device |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JP6996368B2 (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11327483B2 (en) * | 2019-09-30 | 2022-05-10 | Irobot Corporation | Image capture devices for autonomous mobile robots and related systems and methods |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5136504B2 (en) | 2009-04-02 | 2013-02-06 | トヨタ自動車株式会社 | Object identification device |
-
2018
- 2018-03-15 JP JP2018047742A patent/JP6996368B2/en active Active
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3343307B1 (en) | Mapping method, localization method, robot system, and robot | |
| RU2017134401A (en) | METHOD AND SYSTEM FOR TRACKING OBJECTS USING SENSOR DATA FUSION IN A PROBABLE MODEL | |
| CN106780608B (en) | Pose information estimation method and device and movable equipment | |
| US9274526B2 (en) | Autonomous vehicle and method of estimating self position of autonomous vehicle | |
| WO2020140047A4 (en) | Distance to obstacle detection in autonomous machine applications | |
| RU2016152159A (en) | HUMAN-LIKE ROBOT WITH ABILITY TO PREVENT COLLISION AND TO RESTORE A TRAJECTORY | |
| US20110052043A1 (en) | Method of mobile platform detecting and tracking dynamic objects and computer-readable medium thereof | |
| GB2545053A (en) | Collision avoidance using auditory data augmented with map data | |
| EP3338248A1 (en) | Systems and methods for object tracking | |
| JP2016536668A5 (en) | ||
| US10929986B2 (en) | Techniques for using a simple neural network model and standard camera for image detection in autonomous driving | |
| RU2017134760A (en) | PROBABLE IDENTIFICATION OF THE OBJECT BY NUMEROUS SENSORS AND AUTOMATIC BRAKING | |
| JP2010108433A (en) | Autonomous mobile device and control method | |
| US11904853B2 (en) | Apparatus for preventing vehicle collision and method thereof | |
| WO2017038012A1 (en) | Mapping method, localization method, robot system, and robot | |
| WO2018052714A2 (en) | Video to radar | |
| CN115609594B (en) | Planning method and device for mechanical arm path, upper control end and storage medium | |
| US20150183465A1 (en) | Vehicle assistance device and method | |
| WO2023051333A1 (en) | Robot control method, robot, control terminal, and control system | |
| CN116645396A (en) | Track determination method, track determination device, computer-readable storage medium and electronic device | |
| JP2019160003A5 (en) | Obstacle recognition system, server, obstacle recognition device, and database construction method | |
| CN107430821B (en) | Image processing apparatus | |
| US20220017106A1 (en) | Moving object control device, moving object control learning device, and moving object control method | |
| JP2012118683A (en) | Pedestrian recognition device | |
| JP6905433B2 (en) | Vehicle behavior prediction device, vehicle behavior prediction method, and neural network learning method for vehicle behavior prediction |