WO2015155874A1 - 経路予測装置 - Google Patents
経路予測装置 Download PDFInfo
- Publication number
- WO2015155874A1 WO2015155874A1 PCT/JP2014/060427 JP2014060427W WO2015155874A1 WO 2015155874 A1 WO2015155874 A1 WO 2015155874A1 JP 2014060427 W JP2014060427 W JP 2014060427W WO 2015155874 A1 WO2015155874 A1 WO 2015155874A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- collision
- route
- unit
- target object
- value
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G9/00—Traffic control systems for craft where the kind of craft is irrelevant or unspecified
- G08G9/02—Anti-collision systems
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
- G08G1/166—Anti-collision systems for active traffic, e.g. moving vehicles, pedestrians, bikes
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
- G08G1/167—Driving aids for lane monitoring, lane changing, e.g. blind spot detection
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G3/00—Traffic control systems for marine craft
- G08G3/02—Anti-collision systems
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G5/00—Traffic control systems for aircraft, e.g. air-traffic control [ATC]
- G08G5/04—Anti-collision systems
- G08G5/045—Navigation or guidance aids, e.g. determination of anti-collision manoeuvers
Definitions
- the present invention observes the position of a moving target object such as an aircraft, a ship, or a vehicle using an observation device including a sensor such as a radar or a GPS, and based on the observed value, the target object and a plurality of objects existing around the target object are observed.
- the present invention relates to a route prediction device that predicts a route for preventing a collision with an object.
- a vehicle driving support system for a vehicle
- the position of an obstacle such as a vehicle or a stop existing around the own vehicle is acquired by a sensor such as a millimeter wave radar or a laser radar mounted on the own vehicle, and the relative position between the own vehicle and the obstacle is obtained.
- Technology has been developed to prevent collisions by controlling the host vehicle after judging the danger of collision from the distance and relative speed.
- automatic driving technology that recognizes the surrounding environment by the above-mentioned sensors, automatically operates the steering wheel and brakes without the driver's operation, and reaches the destination is being developed. Yes.
- a conventional technique related to such route prediction for example, in the apparatus shown in Patent Document 1, a plurality of vehicle predicted trajectories are generated in advance, and the existence probability of a predicted route in space-time is calculated based on the generated predicted trajectory. . Further, for example, in the driving support device as shown in Patent Document 2, the risk potential map of the own vehicle with respect to another vehicle is calculated, and control of accelerator, brake, etc. based on the risk is made to intervene.
- Patent Document 3 calculates the future position on the assumption that the future position is straight ahead at a constant speed based on the current target speed and the nose direction.
- an optimal route search method using an A * algorithm is used as a method for predicting a future position.
- a node from a departure to a goal (or a waypoint) on a moving space obtained by subdividing a route candidate into a mesh including an entry prohibition area (obstacle) is determined.
- the conventional apparatus as described in Patent Document 1 has a problem in that a large number of predicted trajectories must be generated in order to calculate the existence probability, resulting in a large calculation load.
- the apparatus as disclosed in Patent Document 2 has a problem that the risk calculation method is not clear and is a parameter-dependent calculation method, so that the risk cannot be accurately evaluated.
- the estimation accuracy of the future position is deteriorated when the target changes the course in order to avoid an obstacle such as a thundercloud.
- the system using the A * algorithm as described in Patent Document 4 since the path is determined by the grid points, there is a problem that the movement of the moving body is not considered. In order to obtain a natural path, it is necessary to make the interval between the lattice points fine, and there is a problem that processing time is sacrificed.
- the present invention has been made to solve such a problem, and it is an object of the present invention to provide a route prediction device capable of reducing a calculation load when calculating a prediction route having a low collision risk.
- the route prediction device performs a tracking process based on the position of the target object and the peripheral object, the sensor unit for observing the position of the target object and the peripheral object located around the target object, and the target object and the peripheral object Based on a tracking processing unit that calculates the estimated position and speed of the object, a collision object detection unit that detects a peripheral object that may collide with the target object from the estimated position and speed, and a collision avoidance model A path prediction unit that predicts the path of the target object relative to the target object, a collision risk evaluation unit that calculates the collision risk between the target object and the target object corresponding to the collision avoidance model, and whether or not there is a collision based on the collision risk A collision determination unit that feeds back a collision avoidance model correction value to the route prediction unit, and a plurality of collision times that the collision determination unit determines that there is no collision.
- An avoidance route selection unit that selects one of the collision avoidance models from the model, and selects a route of the collision avoidance model as a route that avoids collision between objects, and the route prediction unit includes a collision avoidance model correction value In this way, new route prediction is performed.
- the path predicting apparatus predicts the path of the target object relative to the target object based on the collision avoidance model, calculates a collision risk between the target object and the target object corresponding to the collision avoidance model, and Judgment of the presence or absence of a collision based on the degree of danger, and one of the collision avoidance models selected from the multiple collision avoidance models determined to have no collision is selected as a path to avoid collision between objects. is there. Thereby, the calculation load at the time of calculating the prediction route with a low collision risk can be reduced.
- FIG. 1 is a configuration diagram showing a route prediction apparatus according to the present embodiment.
- the route prediction apparatus includes a sensor unit 1, a tracking processing unit 2, a collision object detection unit 3, a route prediction unit 4, a collision risk evaluation unit 5, a collision determination unit 6, and a collision avoidance route.
- a selection unit 7 is provided.
- the sensor unit 1 is a processing unit that observes a relative position between a target object and a peripheral object located around the target object, and is a sensor such as a millimeter wave radar, a laser radar, an optical camera, or an infrared camera, or It is comprised using the communication apparatus etc. which receive the GPS position of a surrounding vehicle or a pedestrian.
- the tracking processing unit 2 is a processing unit that performs tracking processing based on the relative position observed by the sensor unit 1 and calculates the estimated position, estimated speed, estimated error of the estimated position, and estimated error of the estimated speed of the target object and the surrounding objects. is there.
- the collision object detection unit 3 is a processing unit that detects, as a target object, a peripheral object that may collide with the target object from the estimated position and the estimated speed.
- the route prediction unit 4 is a processing unit that calculates a predicted position of the target object up to N steps ahead of the target object in each of the M collision avoidance models (where M and N are arbitrary integers).
- the collision risk evaluation unit 5 is a processing unit that calculates the collision risk for each collision avoidance model from the estimated position and the estimation error calculated by the tracking processing unit 2.
- the collision determination unit 6 determines the presence or absence of a collision from the collision risk calculated by the collision risk evaluation unit 5, and when it determines that there is a collision, it feeds back the collision avoidance model correction value to the route prediction unit 4 and determines that there is no collision.
- the processing unit outputs the collision avoidance model to the collision avoidance route selection unit 7.
- the collision avoidance path selection unit 7 selects one of the collision avoidance models based on a predetermined determination criterion for the collision avoidance model output from the collision determination unit 6 and determines a prediction path for collision avoidance Part.
- the route prediction device is configured using a computer, and the tracking processing unit 2 to the collision avoidance route selection unit 7 are realized by executing software corresponding to the functions of the respective processing units on the CPU.
- the sensor unit 1 to the collision avoidance route selection unit 7 may be configured with dedicated hardware.
- the sensor unit 1 measures the position and speed of surrounding vehicles and pedestrians.
- the tracking processing unit 2 calculates a position estimation value, a speed estimation value, and a position and speed estimation error covariance matrix based on the position and the speed through the tracking process.
- the collision object detection unit 3 detects surrounding vehicles that may collide with the host vehicle. For example, it may be detected based on the concept of TTC (Time To Collision). TTC is defined by Formula (1), and if TTC is below a threshold, it will detect as a vehicle which may collide. Further, the detected peripheral vehicle i is defined as the target vehicle.
- TTC Time To Collision
- a predetermined area is set around the own vehicle, and a vehicle in which the predicted position after 1 to N steps enters the predetermined area is detected and regarded as a target vehicle. good.
- N predicted positions up to N steps ahead are calculated as in equation (2).
- the route prediction unit 4 calculates a predicted position up to N steps ahead in each of the M collision avoidance models with respect to the target vehicle tgti detected by the collision object detection unit 3.
- a braking avoidance model, a left steering avoidance model, and a right steering avoidance model may be defined as the collision avoidance model.
- the braking avoidance model is a model that avoids a collision by braking while maintaining the lane
- the left / right steering avoidance model is a model that changes a lane to the left / right by inputting a steering amount to avoid a collision.
- the braking amount or the steering amount is set so as not to exceed a predetermined limit value.
- a corrected value of the braking amount or the steering amount is fed back to the route prediction unit 4, but an operation that does not exceed a predetermined limit value is performed. carry out.
- the route prediction unit 4 needs to set an initial value of the braking amount or the steering amount of the collision avoidance model.
- an initial value a value input at the time of braking and steering avoidance may be set empirically.
- the braking amount and the steering amount may be set so that each driver does not feel uncomfortable using a learning algorithm.
- the route prediction unit 4 is not limited to the above model, and other collision avoidance models corresponding to various scenes may be added.
- unnecessary collision avoidance models may be rejected to reduce the number of collision avoidance models.
- the number of lanes is two and the host vehicle is traveling in the left lane, it is impossible to avoid left steering, so the left steering avoidance model is rejected and the remaining collision avoidance models are calculated.
- map data such as adding a collision avoidance model to change lanes to an additional lane It becomes possible. If a laser radar, camera, or the like is used in addition to the map data, the external environment can be recognized and may be used instead of the map.
- Equation (6) A predicted position calculation method based on the collision avoidance model will be described. Based on the braking acceleration ab of the braking avoidance model, a predicted route (predicted position up to N steps ahead) is calculated as shown in Equation (6).
- the predicted position of the vehicle with respect to steering differs depending on vehicle parameters such as the vehicle weight, the position of the center of gravity of the vehicle body, and the yaw moment of inertia
- the predicted position is calculated in advance.
- a parameter estimated by a known learning algorithm or the like may be used.
- the collision risk evaluation unit 5 defines the collision risk as an upper probability of the chi-square distribution (shaded portion 100 in FIG. 2) as shown in FIG.
- the relative position of the host vehicle (target 2) and the target vehicle (target 1) and the collision risk will be described. For example, in a scene where target 1 and target 2 collide as shown in FIG. 3 (the positions of target 1 and target 2 are the same), the shaded portion 101 in FIG.
- the collision risk is calculated as 1 (or 100%).
- the shaded portion in FIG. That is, the collision risk is calculated as 0 (0%).
- the upper probability of the chi-square distribution is intuitively a value corresponding to the collision risk.
- the correspondence table of the square value ⁇ k + n of the Mahalanobis distance and the upper probability of the chi-square distribution can be calculated in advance, if the correspondence table is retained, the collision risk corresponding to the square value of the Mahalanobis distance can be obtained. Reading without calculation is possible.
- the method for calculating the collision risk at the relative position between the own vehicle and the surrounding vehicle has been described so far.
- a method for calculating the collision risk when the absolute positions of the target 1 and the target 2 are used will be described.
- absolute values such as GPS positions of the own vehicle and surrounding vehicles are acquired by inter-vehicle communication or the like.
- the radar observation position and GPS position are acquired for a plurality of aircraft and used for aircraft control.
- an evaluation value of the collision risk is calculated using the following formulas (12) and (13), and the collision risk corresponding to the evaluation value is read out. .
- the probability distribution of the square value ⁇ k + n of the Mahalanobis distance may be approximated by another probability distribution (for example, a normal distribution).
- the collision determination unit 6 determines a collision based on the collision risk calculated by the collision risk evaluation unit 5, and outputs a predicted route correction value to the route prediction unit 4 in the case of a collision to recorrect the predicted route.
- the predicted route and the collision risk level are output to the collision avoidance route selection unit 7.
- the threshold ⁇ th uses a chi-square distribution table with m degrees of freedom, and as described in the collision risk evaluation unit 5, if the collision threshold ⁇ th corresponding to the collision risk is set in advance, it is easy. It can be determined whether or not it collides.
- the collision risk with 201 and the nearest rear vehicle 202 is calculated. Further, the maximum value is selected from the collision risk levels of the target vehicle 203, the nearest forward vehicle 201, and the nearest rear vehicle 202, and a collision determination is performed. In addition, the area
- the collision determination unit 6 feeds back the corrected value of the predicted route to the route prediction unit 4, whereby the route prediction unit 4 and the collision risk evaluation unit 5 recalculate the predicted route and the collision risk. These procedures are repeated until the threshold ⁇ th is exceeded.
- the processing flow of these route prediction unit 4 to collision determination unit 6 is shown in FIG. That is, for each target vehicle, N-step route prediction (step ST1), collision risk evaluation (step ST2), and collision determination (steps ST3 and ST4) are performed for all models. If the collision threshold value is equal to or less than the collision threshold value in step ST4, the model loop is performed until the collision threshold value is exceeded. If the model loop reaches a predetermined number of times, the calculation of the collision avoidance model may be aborted.
- the collision avoidance route selection unit 7 determines a predicted route for collision avoidance from the predicted routes based on the respective collision avoidance models calculated by the route prediction unit 4 to the collision determination unit 6. For the N predicted positions based on each collision avoidance model, the maximum value of the collision risk is compared, the collision avoidance model having the smallest value is regarded as the safest avoidance route, and is output as the prediction route for collision avoidance. Note that a collision avoidance model that is equal to or smaller than a set value including the minimum value may be selected. Further, a route that minimizes the total of N collision risk levels given to the N predicted positions may be selected. In this case as well, a route that is equal to or less than the set value including the minimum value may be selected. Further, when the braking amount or the steering amount exceeds a predetermined limit value, it may be rejected. Further, a route that minimizes the total braking amount or a route that minimizes the total steering avoidance amount may be selected according to the driver's needs.
- the collision avoidance model As described above, in the first embodiment, by limiting to the collision avoidance model that is realistically assumed, it is not necessary to calculate an infinite number of routes as in the conventional case, and the calculation load can be reduced.
- the route prediction device of the first embodiment based on the sensor unit that observes the position of the target object and the peripheral object located around the target object, and the position of the target object and the peripheral object A tracking processing unit that performs tracking processing to calculate the estimated position and estimated speed of the target object and surrounding objects, and a collision object that detects a surrounding object that may collide with the target object from the estimated position and estimated speed as the target object A detection unit, a route prediction unit that predicts the path of the target object relative to the target object based on the collision avoidance model, and a collision risk evaluation unit that calculates the collision risk between the target object and the target object corresponding to the collision avoidance model
- the collision determination unit determines whether or not there is a collision based on the collision risk level.
- the collision determination unit that feeds back the collision avoidance model correction value to the route prediction unit and the collision determination unit
- a avoidance route selection unit that selects one of the collision avoidance models from the plurality of collision avoidance models determined to be, and selects a route of the collision avoidance model as a route that avoids collision between objects, and a route prediction unit Since the new route prediction is performed using the collision avoidance model correction value, it is possible to reduce the calculation load when calculating a predicted route with a low collision risk.
- the tracking processing unit calculates the estimation error of the estimated position and the estimation error of the estimated speed, and the collision risk evaluation unit normalizes the estimated position with the estimation error. Since the collision risk is calculated from the calculated value, the collision risk can be calculated without requiring complicated numerical calculation.
- the collision risk evaluation unit acquires the collision risk from the correspondence table indicating the correspondence between the value obtained by normalizing the estimated position with the estimation error and the collision risk. Therefore, it is possible to easily obtain the collision risk without requiring numerical calculation.
- the avoidance route selection unit selects a collision avoidance model in which the accumulated value is equal to or less than a set value for the time direction accumulated value of the collision risk of the collision avoidance model. Therefore, it is not necessary to calculate an infinite number of routes, and the calculation load can be reduced.
- the avoidance route selection unit selects a collision avoidance model in which the representative value is equal to or less than the set value with the maximum value in the time direction of the collision risk of the collision avoidance model as a representative value. Since the selection is made, it is not necessary to calculate an infinite number of routes, and the calculation load can be reduced.
- the collision determination unit performs the collision determination by comparing the collision risk with the set threshold value, and thus determines whether or not the collision easily occurs. can do.
- any component of the embodiment can be modified or any component of the embodiment can be omitted within the scope of the invention.
- the route predicting apparatus uses an observation device including sensors such as radar and GPS to observe the position of a moving body such as an aircraft, a ship, and a vehicle, and moves based on the observed value.
- the present invention relates to a route predicting device for predicting a route for preventing a collision between a body and a plurality of moving objects existing in the vicinity thereof, and is suitable for use in a vehicle driving support system or air traffic control.
Abstract
Description
例えば車両の運転支援システムでは、自車両周辺に存在する車両や停止物等の障害物の位置を自車両に搭載したミリ波レーダやレーザレーダ等のセンサによって取得し、自車両と障害物の相対距離や相対速度より衝突危険性を判断した上で自車両を制御し、衝突を防止する技術が開発されている。また、より高度な技術として、前記のセンサによって周辺環境を認識し、運転者の操作無く、自動でハンドル操作やブレーキなどの操作を行い、目的地まで到達する自動運転技術も開発が進められている。
実施の形態1.
図1は、本実施の形態による経路予測装置を示す構成図である。
本実施の形態による経路予測装置は、図示のように、センサ部1、追尾処理部2、衝突物体検知部3、経路予測部4、衝突危険度評価部5、衝突判定部6、衝突回避経路選択部7を備えている。
センサ部1は、周辺車両や歩行者の位置と速度を測定する。追尾処理部2は、位置と速度に基づき、追尾処理を介して位置推定値、速度推定値、位置及び速度の推定誤差共分散行列を算出する。
衝突物体検知部3では、自車両と衝突する可能性のある周辺車両を検知する。例えば、TTC(Time To Collision)の考え方に基づいて検知しても良いこととする。TTCは式(1)で定義され、TTCが閾値以下であれば衝突する可能性がある車両として検知する。さらに、検知された周辺車両iをターゲット車両として定義する。
ここで、衝突回避モデルとして、例えば、制動回避モデル、左操舵回避モデル、右操舵回避モデルを定義しても良い。制動回避モデルは、車線維持したまま制動で衝突を回避するモデル、左/右操舵回避モデルは操舵量を入力することで左/右へ車線変更し、衝突を回避するモデルとする。また、これらのモデルについて、制動量もしくは操舵量は所定の限界値を超えないように設定することとする。特に、後述する衝突判定部6において、衝突回避不可能と判定された場合、制動量もしくは操舵量の修正値が経路予測部4へフィードバックされるが、所定の限界値を超えないような動作を実施する。
また、経路予測部4では、衝突回避モデルの制動量もしくは操舵量の初期値を設定する必要がある。初期値として、制動および操舵回避の際に入力される値を経験的に設定しても良い。または、学習アルゴリズムを利用して、運転者ごとに不快に感じない程度の制動量、操舵量を設定しても良い。
式(9)のようにサンプリング時刻kにおける自車両のnステップ後の予測位置とターゲット車両tgtiのn(n=1,…,N)ステップ後の予測位置の差分を推定誤差共分散行列で正規化した値、つまりマハラノビス距離の2乗値εk+nを算出する。
衝突危険度を直感的に理解するために、自車両(目標2)とターゲット車両(目標1)の相対位置と衝突危険度の対応関係について説明する。例えば、図3のように目標1と目標2が衝突するシーン(目標1と目標2の位置が同じ)では、図3の斜線部分101が1に近づく。つまり、衝突危険度は1(もしくは100%)として算出される。一方、図4のように目標1と目標2との距離が限りなく遠いシーンでは、図4の斜線部分は0に近づく。つまり、衝突危険度は0(0%)として算出される。そのため、カイ二乗分布の上側確率は、直感的にも衝突危険度に相当する値であることがわかる。さらに、マハラノビス距離の2乗値εk+nとカイ二乗分布の上側確率の対応表は予め計算可能であるため、対応表を保持しておけば、マハラノビス距離の2乗値に応じた衝突危険度を計算無く読み出すことが可能となる。
また、マハラノビス距離の2乗値εk+nの確率分布を他の確率分布(例えば正規分布)で近似しても良い。
衝突判定には確率変数εk+n(n=1,…,N)の最小値に対して閾値εth以下である場合は衝突することとみなす。前記の閾値εthは自由度mのカイ二乗分布表を用いることとし、衝突危険度評価部5でも述べたように、予め衝突危険度に対応する衝突閾値εthを設定しておけば、容易に衝突するか否かを判定できる。
これら経路予測部4~衝突判定部6の処理フローを図6に示す。すなわち、ターゲット車両ごとに、全てのモデルに対してNステップの経路予測(ステップST1)と衝突危険度の評価(ステップST2)および衝突判定(ステップST3,ST4)を行う。また、ステップST4で衝突閾値以下である場合は、衝突閾値を超えるまでモデルループを行う。なお、モデルループが予め定めた所定回数に達した場合は、その衝突回避モデルの計算を打ち切るようにしてもよい。
各衝突回避モデルに基づくN個の予測位置について、衝突危険度の最大値を比較し、最も小さい値を持つ衝突回避モデルを最も安全な回避経路とみなし、衝突回避用予測経路として出力する。なお、最小値を含む設定値以下の衝突回避モデルを選択するようにしてもよい。
また、前記N個の予測位置に付与されるN個の衝突危険度の合計を比較して最小となる経路を選択しても良い。なお、ここでも最小値を含む設定値以下の経路を選択するようにしてもよい。
また、制動量もしくは操舵量が所定の限界値を超える場合は、棄却しても良い。
また、運転者のニーズに合わせて、制動量の合計値が最小となる経路、もしくは操舵回避量の合計値が最小となる経路を選択しても良い。
Claims (6)
- 対象物体と当該対象物体の周辺に位置する周辺物体との位置を観測するセンサ部と、
前記対象物体と前記周辺物体の位置に基づいて追尾処理を行い、前記対象物体と前記周辺物体の推定位置と推定速度を算出する追尾処理部と、
前記推定位置と前記推定速度から、前記対象物体と衝突する可能性のある周辺物体をターゲット物体として検知する衝突物体検知部と、
衝突回避モデルに基づき、前記ターゲット物体に対する前記対象物体の経路を予測する経路予測部と、
前記対象物体と前記ターゲット物体との衝突危険度を前記衝突回避モデルに対応して算出する衝突危険度評価部と、
前記衝突危険度より衝突の有無を判定し、衝突と判定した場合は前記経路予測部へ衝突回避モデル修正値をフィードバックする衝突判定部と、
前記衝突判定部で衝突無しと判定された複数の衝突回避モデルから、いずれかの衝突回避モデルを選択し、当該衝突回避モデルの経路を、物体同士の衝突を回避する経路として選択する回避経路選択部とを備え、
前記経路予測部は、前記衝突回避モデル修正値を用いて、新たな経路予測を行うことを特徴とする経路予測装置。 - 前記追尾処理部は、前記推定位置の推定誤差と前記推定速度の推定誤差とを算出し、
前記衝突危険度評価部は、前記推定位置を前記推定誤差で正規化した値より衝突危険度を算出することを特徴とする請求項1記載の経路予測装置。 - 前記衝突危険度評価部は、前記推定位置を前記推定誤差で正規化した値と衝突危険度との対応を示す対応表より衝突危険度を取得することを特徴とする請求項1記載の経路予測装置。
- 前記回避経路選択部は、前記衝突回避モデルの衝突危険度の時間方向累積値について、当該累積値が設定値以下となる衝突回避モデルを選択することを特徴とする請求項1記載の経路予測装置。
- 前記回避経路選択部は、前記衝突回避モデルの衝突危険度の時間方向に対する最大値を代表値として、当該代表値が設定値以下となる衝突回避モデルを選択することを特徴とする請求項1記載の経路予測装置。
- 前記衝突判定部は、前記衝突危険度を設定された閾値と比較することにより衝突判定を行うことを特徴とする請求項1記載の経路予測装置。
Priority Applications (5)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
DE112014006561.7T DE112014006561T5 (de) | 2014-04-10 | 2014-04-10 | Routenvorausberechunungseinrichtung |
JP2016512545A JP6203381B2 (ja) | 2014-04-10 | 2014-04-10 | 経路予測装置 |
PCT/JP2014/060427 WO2015155874A1 (ja) | 2014-04-10 | 2014-04-10 | 経路予測装置 |
US15/129,138 US10102761B2 (en) | 2014-04-10 | 2014-04-10 | Route prediction device |
CN201480077864.1A CN106164998B (zh) | 2014-04-10 | 2014-04-10 | 路径预测装置 |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2014/060427 WO2015155874A1 (ja) | 2014-04-10 | 2014-04-10 | 経路予測装置 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2015155874A1 true WO2015155874A1 (ja) | 2015-10-15 |
Family
ID=54287475
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2014/060427 WO2015155874A1 (ja) | 2014-04-10 | 2014-04-10 | 経路予測装置 |
Country Status (5)
Country | Link |
---|---|
US (1) | US10102761B2 (ja) |
JP (1) | JP6203381B2 (ja) |
CN (1) | CN106164998B (ja) |
DE (1) | DE112014006561T5 (ja) |
WO (1) | WO2015155874A1 (ja) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2017142145A (ja) * | 2016-02-10 | 2017-08-17 | 株式会社ゼンリン | 車線変更支援装置、車両制御装置およびその方法 |
JP2019053633A (ja) * | 2017-09-19 | 2019-04-04 | トヨタ自動車株式会社 | 車両周辺監視装置 |
WO2022162996A1 (ja) * | 2021-01-27 | 2022-08-04 | 日立Astemo株式会社 | 電子制御装置および車両制御システム |
US11417219B2 (en) | 2017-04-20 | 2022-08-16 | Fujitsu Limited | Non-transitory computer-readable storage medium for storing collision risk calculation program, collision risk calculation method, and collision risk calculation apparatus |
Families Citing this family (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP6234635B2 (ja) * | 2015-04-23 | 2017-11-22 | 三菱電機株式会社 | 評価情報収集装置 |
KR101714273B1 (ko) * | 2015-12-11 | 2017-03-08 | 현대자동차주식회사 | 자율 주행 시스템의 경로 제어 방법 및 그 장치 |
US9766344B2 (en) * | 2015-12-22 | 2017-09-19 | Honda Motor Co., Ltd. | Multipath error correction |
JP6603197B2 (ja) * | 2016-11-29 | 2019-11-06 | 株式会社デンソー | 連続値最適化問題の非線形最適化プログラム、経路探索プログラム、及び経路探索装置 |
JP6616275B2 (ja) * | 2016-12-15 | 2019-12-04 | 株式会社Soken | 運転支援装置 |
KR102215325B1 (ko) * | 2017-02-28 | 2021-02-15 | 현대자동차주식회사 | 차량의 위치 추정 장치 및 방법과 이를 이용한 차량 |
JP6652106B2 (ja) * | 2017-05-10 | 2020-02-19 | 株式会社ダイフク | 物品搬送設備 |
EP3413082B1 (en) * | 2017-06-09 | 2020-01-01 | Veoneer Sweden AB | A vehicle system for detection of oncoming vehicles |
KR20190014871A (ko) * | 2017-08-04 | 2019-02-13 | 삼성전자주식회사 | 차량의 후방에 위치한 응급 차량에 기초하여 상기 차량의 주행 경로를 변경하는 장치 및 방법 |
CN110276985B (zh) * | 2018-03-16 | 2020-12-15 | 华为技术有限公司 | 自动驾驶安全评估方法、装置和系统 |
JP7111517B2 (ja) * | 2018-06-14 | 2022-08-02 | シャープ株式会社 | 走行装置、走行装置の走行制御方法、走行装置の走行制御プログラムおよび記録媒体 |
CN111328411B (zh) * | 2018-09-28 | 2022-11-29 | 百度时代网络技术(北京)有限公司 | 用于自动驾驶车辆的行人概率预测系统 |
US20220227395A1 (en) * | 2019-05-23 | 2022-07-21 | Hitachi Astemo, Ltd. | Vehicle Control System and Vehicle Control Method |
JP7232166B2 (ja) * | 2019-10-18 | 2023-03-02 | 本田技研工業株式会社 | 予測装置、車両システム、予測方法、およびプログラム |
EP4059003A4 (en) * | 2019-11-13 | 2023-11-22 | Zoox, Inc. | COLLISION MONITORING USING STATISTICAL MODELS |
US10971005B1 (en) * | 2019-12-26 | 2021-04-06 | Continental Automotive Systems, Inc. | Determining I2X traffic-participant criticality |
US20210339741A1 (en) * | 2020-04-30 | 2021-11-04 | Zoox, Inc. | Constraining vehicle operation based on uncertainty in perception and/or prediction |
TWI811692B (zh) * | 2020-06-12 | 2023-08-11 | 中央研究院 | 用於場景音轉換的方法與裝置及電話系統 |
US20220155082A1 (en) * | 2020-11-13 | 2022-05-19 | Uber Technologies, Inc. | Route comparison for vehicle routing |
CN115394126B (zh) * | 2022-10-25 | 2023-01-20 | 北京开运联合信息技术集团股份有限公司 | 一种空间目标的碰撞预警方法及装置 |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2007276508A (ja) * | 2006-04-03 | 2007-10-25 | Fujitsu Ten Ltd | 車両の衝突回避制御装置 |
JP2008265468A (ja) * | 2007-04-18 | 2008-11-06 | Nissan Motor Co Ltd | 走行支援装置 |
Family Cites Families (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2985952B2 (ja) | 1997-10-20 | 1999-12-06 | 日本電気株式会社 | 航空機接近警報装置 |
DE10319700A1 (de) * | 2003-05-02 | 2004-11-18 | Ibeo Automobile Sensor Gmbh | Verfahren und Vorrichtung zur Ermittlung einer Wahrscheinlichkeit für eine Kollision eines Fahrzeugs mit einem Gegenstand |
US7262710B2 (en) * | 2004-09-22 | 2007-08-28 | Nissan Motor Co., Ltd. | Collision time estimation apparatus for vehicles, collision time estimation method for vehicles, collision alarm apparatus for vehicles, and collision alarm method for vehicles |
WO2007102367A1 (ja) | 2006-02-28 | 2007-09-13 | Toyota Jidosha Kabushiki Kaisha | 物体進路予測方法、装置、プログラム、および自動運転システム |
JP4396653B2 (ja) | 2006-02-28 | 2010-01-13 | トヨタ自動車株式会社 | 物体進路予測方法、装置、およびプログラム |
JP4400634B2 (ja) * | 2007-02-28 | 2010-01-20 | トヨタ自動車株式会社 | 衝突予測装置 |
JP5250290B2 (ja) | 2008-04-02 | 2013-07-31 | 富士重工業株式会社 | 航空機の四次元最適経路誘導システム |
KR101619076B1 (ko) * | 2009-08-25 | 2016-05-10 | 삼성전자 주식회사 | 모바일 플랫폼의 동적 물체 검출 및 추적 방법 |
US8655579B2 (en) * | 2010-03-16 | 2014-02-18 | Toyota Jidosha Kabushiki Kaisha | Driving assistance device |
US8954260B2 (en) * | 2010-06-15 | 2015-02-10 | GM Global Technology Operations LLC | Method and system for collision assessment for vehicles |
JP2013532858A (ja) * | 2010-07-17 | 2013-08-19 | ヴァレオ・シャルター・ウント・ゼンゾーレン・ゲーエムベーハー | 自動車の側面に隣接する側面域に存在する障害物について自動車の運転者に警告する方法、および運転者支援システムを備える自動車 |
JP5673127B2 (ja) | 2011-01-21 | 2015-02-18 | トヨタ自動車株式会社 | 運転支援装置 |
FR2988507B1 (fr) * | 2012-03-23 | 2014-04-25 | Inst Francais Des Sciences Et Technologies Des Transports De Lamenagement Et Des Reseaux | Systeme d'assistance pour vehicule routier |
-
2014
- 2014-04-10 US US15/129,138 patent/US10102761B2/en active Active
- 2014-04-10 WO PCT/JP2014/060427 patent/WO2015155874A1/ja active Application Filing
- 2014-04-10 CN CN201480077864.1A patent/CN106164998B/zh active Active
- 2014-04-10 DE DE112014006561.7T patent/DE112014006561T5/de active Pending
- 2014-04-10 JP JP2016512545A patent/JP6203381B2/ja active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2007276508A (ja) * | 2006-04-03 | 2007-10-25 | Fujitsu Ten Ltd | 車両の衝突回避制御装置 |
JP2008265468A (ja) * | 2007-04-18 | 2008-11-06 | Nissan Motor Co Ltd | 走行支援装置 |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2017142145A (ja) * | 2016-02-10 | 2017-08-17 | 株式会社ゼンリン | 車線変更支援装置、車両制御装置およびその方法 |
US11417219B2 (en) | 2017-04-20 | 2022-08-16 | Fujitsu Limited | Non-transitory computer-readable storage medium for storing collision risk calculation program, collision risk calculation method, and collision risk calculation apparatus |
JP2019053633A (ja) * | 2017-09-19 | 2019-04-04 | トヨタ自動車株式会社 | 車両周辺監視装置 |
WO2022162996A1 (ja) * | 2021-01-27 | 2022-08-04 | 日立Astemo株式会社 | 電子制御装置および車両制御システム |
Also Published As
Publication number | Publication date |
---|---|
DE112014006561T5 (de) | 2017-02-16 |
CN106164998A (zh) | 2016-11-23 |
JPWO2015155874A1 (ja) | 2017-04-13 |
CN106164998B (zh) | 2019-03-15 |
US20170039865A1 (en) | 2017-02-09 |
US10102761B2 (en) | 2018-10-16 |
JP6203381B2 (ja) | 2017-09-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP6203381B2 (ja) | 経路予測装置 | |
US11645916B2 (en) | Moving body behavior prediction device and moving body behavior prediction method | |
JP6800899B2 (ja) | 視界に制限のある交差点への接近のためのリスクベースの運転者支援 | |
CN108137051B (zh) | 用于控制自身车辆的路径规划的方法、装置以及车辆 | |
EP3091370B1 (en) | Method and arrangement for determining safe vehicle trajectories | |
JP5167051B2 (ja) | 車両の運転支援装置 | |
KR102128567B1 (ko) | 제동을 위한 트리거링 기준을 결정하기 위한 방법 및 차량용 비상 제동 시스템 | |
EP3686864A1 (en) | Information processing device, autonomous moving apparatus, method, and program | |
EP2654028B1 (en) | Orientation sensitive traffic collision warning system | |
JP2016051465A (ja) | 衝突後操縦計画を行う方法及びシステム、並びに当該システムを備える車両 | |
JP7156394B2 (ja) | 他車動作予測方法及び他車動作予測装置 | |
Kim et al. | Probabilistic threat assessment with environment description and rule-based multi-traffic prediction for integrated risk management system | |
JP2008171207A (ja) | 車両の運転支援装置 | |
JP5577126B2 (ja) | 走行支援装置 | |
CN113460081A (zh) | 车辆控制装置、车辆控制方法及存储介质 | |
JP2014232508A (ja) | 回避軌道予測装置 | |
WO2020161512A1 (ja) | 他車動作予測方法及び他車動作予測装置 | |
KR20190045308A (ko) | 차량 판정 방법, 주행 경로 보정 방법, 차량 판정 장치, 및 주행 경로 보정 장치 | |
US20210253136A1 (en) | Vehicle control device, vehicle control method, and storage medium | |
WO2019034514A1 (en) | METHOD AND SYSTEM FOR AVOIDING THE COLLISION OF A VEHICLE | |
US20220105963A1 (en) | Systems and methods for imminent collision avoidance | |
JP2019153025A (ja) | 車両制御装置 | |
ES2766148A1 (es) | Metodo y sistema para controlar una pluralidad de vehiculos autonomos | |
WO2021089608A1 (en) | Adaptive cruise control | |
JP7356892B2 (ja) | 車両の走行環境推定方法、及び、走行環境推定システム |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 14888574 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2016512545 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 15129138 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: 112014006561 Country of ref document: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 14888574 Country of ref document: EP Kind code of ref document: A1 |