JP2011221668A - Collision time calculation device - Google Patents

Collision time calculation device Download PDF

Info

Publication number
JP2011221668A
JP2011221668A JP2010088094A JP2010088094A JP2011221668A JP 2011221668 A JP2011221668 A JP 2011221668A JP 2010088094 A JP2010088094 A JP 2010088094A JP 2010088094 A JP2010088094 A JP 2010088094A JP 2011221668 A JP2011221668 A JP 2011221668A
Authority
JP
Japan
Prior art keywords
collision
time
ttc
probability
collision time
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
Application number
JP2010088094A
Other languages
Japanese (ja)
Other versions
JP5607410B2 (en
Inventor
Shinichi Nagata
真一 永田
Masayuki Shimizu
政行 清水
Akio Fukamachi
映夫 深町
Jun Sakukawa
純 佐久川
Tsukasa Shimizu
司 清水
Yoshiko Ohama
吉紘 大濱
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Toyota Motor Corp
Toyota Central R&D Labs Inc
Original Assignee
Toyota Motor Corp
Toyota Central R&D Labs Inc
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Toyota Motor Corp, Toyota Central R&D Labs Inc filed Critical Toyota Motor Corp
Priority to JP2010088094A priority Critical patent/JP5607410B2/en
Publication of JP2011221668A publication Critical patent/JP2011221668A/en
Application granted granted Critical
Publication of JP5607410B2 publication Critical patent/JP5607410B2/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Traffic Control Systems (AREA)

Abstract

PROBLEM TO BE SOLVED: To provide a collision time calculation device capable of calculating a collision time TTC (Time to Collision) including possibility of a collision even in such a case that a collision object does not actually exist.SOLUTION: The collision time TTC is calculated even to a predicted collision scene (a scene where the collision is expressed from the viewpoint of probability) such as dash of a pedestrian from a blind corner. Specifically, a weighted linear sum between an expectation value (a probability average) of the TTC when the collision is predicted by a collision prediction and an expectation value (a preset finite time length) of the TTC when the collision is not predicted is calculated as the collision time to the predicted collision scene.

Description

本発明は衝突時間演算装置に関する。   The present invention relates to a collision time calculation device.

非特許文献1には、先行車追従場面において、“高速度で車間距離が狭く、先行車との相対速度が小さい”という状況では衝突時間TTC(Time to Collision)は非常に大きな値になるので、ある減速度を先行車に設定した場合の衝突時間を新たな危険度評価指標とすることが開示されている。   Non-Patent Document 1 states that in a situation where a preceding vehicle follows, the collision time TTC (Time to Collision) becomes a very large value in a situation where “the speed is high and the inter-vehicle distance is narrow and the relative speed with the preceding vehicle is small”. It is disclosed that the collision time when a certain deceleration is set for the preceding vehicle is used as a new risk evaluation index.

なお、特許文献1には、立体物情報で自車両からの死角となっている位置には予め定めた衝突危険度を設定する車両の運転支援装置に関する技術が開示されている。   Patent Document 1 discloses a technique related to a vehicle driving support device that sets a predetermined collision risk level at a position that is a blind spot from the host vehicle in the three-dimensional object information.

特開2008−171207号公報JP 2008-171207 A

若林拓史ら,“交通流ビデオ解析システムを用いた交通コンフリクト分析と新しい危険度評価指標の提案”,土木計画学研究・論文集,巻:20号,4頁,pp.949−956,2003Wakabayashi Takushi et al., “Traffic Conflict Analysis Using Traffic Video Analysis System and Proposal of New Risk Evaluation Index”, Civil Engineering Planning Research Papers, Volume: 20, No. 4, p. 949-956, 2003

しかしながら、従来技術では、死角から歩行者が飛び出すという予測(実際に衝突対象が顕在化していない状態)等に対しては、衝突時間TTCを算出することができない、という問題点があった。   However, the conventional technique has a problem that the collision time TTC cannot be calculated for a prediction that a pedestrian jumps out of the blind spot (a state where the collision target is not actually revealed).

本発明は、上記の事情に鑑みてなされたものであって、実際に衝突対象がないような場面であっても、衝突の可能性を含めた衝突時間TTCの算出ができる衝突時間演算装置を提供することを目的とする。   The present invention has been made in view of the above circumstances, and a collision time calculation device capable of calculating the collision time TTC including the possibility of a collision even in a scene where there is no actual collision target. The purpose is to provide.

本発明にかかる衝突時間演算装置は、衝突する場合と衝突しない場合とを含めて将来の衝突時間の分布を仮定して将来の衝突時間の期待値を求め、求めた期待値を用いて予測衝突場面に対する衝突時間を演算する衝突時間演算装置において、衝突しない場合の衝突時間を固定値に設定すること、を特徴とする。   The collision time calculation device according to the present invention calculates the expected value of the future collision time by assuming the distribution of the future collision time including the case of collision and the case of no collision, and uses the calculated expected value for the predicted collision. In a collision time calculation device for calculating a collision time for a scene, the collision time when no collision occurs is set to a fixed value.

本発明によれば、衝突する場合と衝突しない場合とを含めて将来の衝突時間の分布を仮定して将来の衝突時間tTTCの期待値を求め、求めた期待値を用いて予測衝突場面に対する衝突時間TTCを演算する際において、衝突しない場合の衝突時間TTCを固定値Tcに設定する。これにより、実際に衝突対象がないような場面であっても、衝突の可能性を含めた衝突時間TTCの算出ができるという効果を奏する。 According to the present invention, the expected value of the future collision time t TTC is obtained by assuming the distribution of the future collision time including the case of collision and the case of no collision, and the predicted value of the predicted collision scene is calculated using the obtained expected value. When calculating the collision time TTC, the collision time TTC when no collision occurs is set to a fixed value Tc. Thereby, there is an effect that the collision time TTC including the possibility of the collision can be calculated even in a scene where there is no actual collision target.

図1は、本実施形態の概要を示す図である。FIG. 1 is a diagram showing an outline of the present embodiment. 図2は、本実施形態の衝突時間演算装置の構成を示すブロック図である。FIG. 2 is a block diagram showing the configuration of the collision time calculation device of the present embodiment. 図3は、本実施形態の衝突時間演算動作の一例を示すフローチャートである。FIG. 3 is a flowchart showing an example of the collision time calculation operation of the present embodiment. 図4は、駐車車両の側方通過時の歩行者飛び出し場面の一例を示す図である。FIG. 4 is a diagram illustrating an example of a pedestrian jumping scene when the parked vehicle passes by the side. 図5は、歩行者位置の分布の一例を示す図である。FIG. 5 is a diagram illustrating an example of the distribution of pedestrian positions. 図6は、歩行者位置の分布の一例を示す図である。FIG. 6 is a diagram illustrating an example of the distribution of pedestrian positions.

以下に、本発明にかかる衝突時間演算装置の実施形態を図面に基づいて説明する。なお、本発明は本実施形態により限定されるものではない。   Hereinafter, embodiments of a collision time calculation device according to the present invention will be described with reference to the drawings. In addition, this invention is not limited by this embodiment.

[1.概要]
ここでは、本実施形態の概要について、図1を参照して説明する。図1は、本実施形態の概要を示す図である。
[1. Overview]
Here, an outline of the present embodiment will be described with reference to FIG. FIG. 1 is a diagram showing an outline of the present embodiment.

本実施形態は、衝突しない場合に対しても衝突時間TTCを定義するためのものである。本実施形態の基本となる考え方は、「衝突する場合も衝突しない場合も含めて、図1に示すような将来の衝突時間tTCCの分布(例えば確率密度分布)を仮定し、将来の衝突時間tTCCの確率平均(期待値)を求める。」というものである。 This embodiment is for defining the collision time TTC even when there is no collision. The basic idea of the present embodiment is that a future collision time t TCC distribution (for example, probability density distribution) as shown in FIG. probability average of t TCC seek the (expected value). "is that.

一般に、衝突しない場合のTTCは無限大となり、それゆえ、概念的には、衝突しない場合のTTCの分布は、無限大の時刻で衝突しない確率分の大きさのインパルスとして考えることができる。   In general, the TTC when there is no collision is infinite. Therefore, conceptually, the distribution of TTC when there is no collision can be considered as an impulse having a magnitude corresponding to the probability of no collision at an infinite time.

この考えは、以下の式(1)で表すことができる。式(1)において、Cは衝突事象(C=0:衝突しない、C=1:衝突)とする。式(1)の第1項は、衝突すると予測される場合の衝突時間の期待値に全時刻に亘っての衝突確率を乗じたものである。   This idea can be expressed by the following formula (1). In Equation (1), C is a collision event (C = 0: no collision, C = 1: collision). The first term of equation (1) is obtained by multiplying the expected value of the collision time when a collision is predicted by the collision probability over all times.

ここで、衝突しないと予測される場合のTTCを、ある有限値をとる分布として定めれば、式(1)の第2項は有限値として求めることができる。そのため、本実施形態では、式(1)の第2項を有限値Tcと設定することによって、式(1)全体を以下の式(2)のように有限値として求めること、を特長とするものである。   Here, if the TTC when predicted not to collide is defined as a distribution having a certain finite value, the second term of the equation (1) can be obtained as a finite value. Therefore, the present embodiment is characterized in that the entire expression (1) is obtained as a finite value as in the following expression (2) by setting the second term of the expression (1) as a finite value Tc. Is.

[2.構成]
ここでは、本実施形態の衝突時間演算装置の構成について図2を参照しながら説明する。図2は、本実施形態の衝突時間演算装置の構成を示すブロック図である。
[2. Constitution]
Here, the configuration of the collision time calculation device of the present embodiment will be described with reference to FIG. FIG. 2 is a block diagram showing the configuration of the collision time calculation device of the present embodiment.

図2において、符号1は、移動体としての車両に搭載されたECU(電子制御ユニット)に組み込まれた衝突時間演算装置である。符号1aは衝突予測部であり、符号1bは衝突時間演算部であり、符号1cは仮想時間設定部であり、符号1dは仮想衝突時間演算部であり、符号1eは重みα設定部であり、符号1fは重みβ設定部であり、符号1gは加算部である。   In FIG. 2, the code | symbol 1 is the collision time calculating device integrated in ECU (electronic control unit) mounted in the vehicle as a moving body. Reference numeral 1a is a collision prediction section, reference numeral 1b is a collision time calculation section, reference numeral 1c is a virtual time setting section, reference numeral 1d is a virtual collision time calculation section, and reference numeral 1e is a weight α setting section. Reference numeral 1f is a weight β setting unit, and reference numeral 1g is an addition unit.

衝突予測部1aは、現在時刻の自車両、他者(他の車両、歩行者など)、道路環境などの情報から、有限な長さの将来時刻Tmaxまでの各将来時刻tにおける衝突確率を求めると共に、将来時刻Tmaxまでの累積衝突確率も求める。 The collision prediction unit 1a calculates the collision probability at each future time t from the information such as the current vehicle at the current time, other persons (other vehicles, pedestrians, etc.), road environment, etc. to a future time Tmax of a finite length. At the same time, the cumulative collision probability up to the future time Tmax is also obtained.

衝突時間演算部1bは、衝突が予測される場合(各将来時刻tにおいて衝突確率が0以上の場合)の衝突時間を、各時刻tの値から求める。なお、当該衝突時間を求める際には、各時刻tにおける衝突確率を用いてもよい。具体的には、衝突時間演算部1bは、時刻tと、時刻tの確率分布を用いて、時刻tの確率平均(期待値)を、衝突が予測される場合の衝突時間として求める。   The collision time calculation unit 1b obtains the collision time when the collision is predicted (when the collision probability is 0 or more at each future time t) from the value at each time t. In addition, when calculating | requiring the said collision time, you may use the collision probability in each time t. Specifically, the collision time calculation unit 1b uses the time t and the probability distribution at the time t to obtain the probability average (expected value) at the time t as the collision time when the collision is predicted.

仮想時間設定部1cは、衝突が予測されなかった場合(各将来時刻tにおいて衝突確率が1未満の場合)に用いる仮想的な衝突時間(仮想時間)Tcを設定する。仮想時間Tcは、スカラー値でもよく、また分布を持つ確率変数でもよい。仮想時間Tcは、自車両の走行速度に応じて変化させて定めてもよい。仮想時間Tcの値は、ユーザが調整してもよい。   The virtual time setting unit 1c sets a virtual collision time (virtual time) Tc used when no collision is predicted (when the collision probability is less than 1 at each future time t). The virtual time Tc may be a scalar value or a random variable having a distribution. The virtual time Tc may be determined by changing according to the traveling speed of the host vehicle. The value of the virtual time Tc may be adjusted by the user.

仮想衝突時間演算部1dは、仮想時間Tcから衝突時間を求める。仮想衝突時間演算部1dは、仮想時間Tcが確率変数である場合は、仮想時間Tcの値として確率平均(期待値)を用いる。   The virtual collision time calculation unit 1d calculates the collision time from the virtual time Tc. When the virtual time Tc is a random variable, the virtual collision time calculation unit 1d uses a probability average (expected value) as the value of the virtual time Tc.

重みα設定部1eは、衝突時間演算部1bで演算された衝突時間に対する重みαを設定する。重みαは、予め決められた固定値でもよく、また衝突確率から求めたものでもよい。具体的には、重みα設定部1eは、衝突予測部1aで求めた将来時刻Tmaxまでの累積衝突確率を重みαの値として用いる。 The weight α setting unit 1e sets the weight α for the collision time calculated by the collision time calculation unit 1b. The weight α may be a predetermined fixed value or may be obtained from the collision probability. Specifically, the weight α setting unit 1e uses the cumulative collision probability up to the future time T max obtained by the collision prediction unit 1a as the value of the weight α.

重みβ設定部1fは、仮想衝突時間演算部1dで演算された衝突時間に対する重みβを設定する。重みβは、予め決められた固定値でもよく、また衝突確率から求めたものでもよい。具体的には、重みβ設定部1fは、1から、衝突予測部1aで求めた将来時刻Tmaxまでの累積衝突確率を引いた値を、重みβの値として用いる。 The weight β setting unit 1f sets the weight β for the collision time calculated by the virtual collision time calculation unit 1d. The weight β may be a predetermined fixed value or may be obtained from the collision probability. Specifically, the weight β setting unit 1 f uses a value obtained by subtracting the cumulative collision probability from 1 to the future time T max obtained by the collision prediction unit 1 a as the value of the weight β.

加算部1gは、衝突時間演算部1bで演算された衝突時間に重みαを乗じた値と、仮想衝突時間演算部1dで演算された衝突時間に重みβを乗じた値とを加算し、それを最終的な衝突時間とする。   The adding unit 1g adds the value obtained by multiplying the collision time calculated by the collision time calculating unit 1b by the weight α and the value obtained by multiplying the collision time calculated by the virtual collision time calculating unit 1d by the weight β, Is the final collision time.

[3.動作]
つぎに、上述した構成の衝突時間演算装置1で行われる衝突時間演算動作の一例について、図3等を参照して説明する。図3は、本実施形態の衝突時間演算動作の一例を示すフローチャートである。ここでは、図4に示すような、駐車車両の側方通過時の歩行者飛び出し場面についての一例を示す。
[3. Operation]
Next, an example of the collision time calculation operation performed by the collision time calculation device 1 having the above-described configuration will be described with reference to FIG. FIG. 3 is a flowchart showing an example of the collision time calculation operation of the present embodiment. Here, an example of a pedestrian jumping scene when a parked vehicle passes sideways as shown in FIG. 4 is shown.

ここで、図4に示す場面について簡単に説明する。時刻t=0において、自車両が地点Xe0(xe0,ye0)を速度V(vxe,vye)で走行中、前方に駐車車両があったとする。図4に示す場面は、従来手法(論文「若林拓史ら,“交通流ビデオ解析システムを用いた交通コンフリクト分析と新しい危険度評価指標の提案”,土木計画学研究・論文集,巻:20号,4頁,pp.949−956,2003」)では、自車両の前方に衝突対象となる歩行者が存在していないのでTTCが無限大となる場面である。そこで、ここでは、駐車車両の死角から歩行者が飛び出すと予測した場合の衝突時間を求める。 Here, the scene shown in FIG. 4 will be briefly described. Assume that at time t = 0, the host vehicle is traveling at a speed V e (v xe , v ye ) at a point X e0 (x e0 , y e0 ) and there is a parked vehicle ahead. The scene shown in FIG. 4 is based on the conventional method (Paper “Takushi Wakabayashi et al.,“ Traffic Conflict Analysis Using Traffic Traffic Video Analysis System and Proposal of New Risk Assessment Index ”, Civil Engineering Planning Research Papers, Volume: No. 20 , Page 4, pp. 949-956, 2003 "), there is no pedestrian to be collided in front of the host vehicle, and the TTC is infinite. Therefore, here, the collision time when the pedestrian is predicted to jump out from the blind spot of the parked vehicle is obtained.

まず、衝突予測部1aは、有限な長さの将来時刻Tmaxまでの各将来時刻tにおける衝突確率Pcol(t)および将来時刻Tmaxまでの累積衝突確率P(C=1)を求める(ステップSA1)。ここで、衝突確率Pcol(t)および累積衝突確率P(C=1)の求め方について、以下に説明する。 First, the collision prediction unit 1a obtains a collision probability P col (t) at each future time t up to a future time T max of a finite length and a cumulative collision probability P (C = 1) up to the future time T max ( Step SA1). Here, how to obtain the collision probability P col (t) and the cumulative collision probability P (C = 1) will be described below.

まず、将来時刻tにおける自車両および歩行者の位置を予測する必要があるが、ここでは自車両は等速で走行するとし、y方向の速度成分のみを持つものとする。この場合、自車両の将来時刻tにおける位置Xetは、以下の式(3)で表される。
et=Xeo+V・t=(xe0,ye0+vyet) ・・・(3)
First, it is necessary to predict the positions of the host vehicle and the pedestrian at a future time t. Here, it is assumed that the host vehicle travels at a constant speed and has only a speed component in the y direction. In this case, the position X et at future time t of the vehicle is expressed by the following equation (3).
X et = X eo + V e · t = (x e0, y e0 + v ye t) ··· (3)

歩行者については、2次元平面上に確率的に分布するものとして考える。駐車車両の死角に歩行者が存在する確率をPとし、歩行者が存在した場合の時刻t=0における歩行者の位置Xp0を2次元正規分布に従うとすると、位置Xp0は以下の式(4)の分布に従う。 A pedestrian is considered to be probabilistically distributed on a two-dimensional plane. If the probability that a pedestrian is present in the blind spot of the parked vehicle is P and the position Xp0 of the pedestrian at time t = 0 when a pedestrian is present follows a two-dimensional normal distribution, the position Xp0 is expressed by Follow the distribution of 4).

ここで、歩行者は速度V(Vxp,Vyp)で等速に歩行するとし、x方向の速度成分のみを持つものとする。この場合、歩行者の時刻tにおける位置Xpt(xpt、ypt)の分布は、XpoをX軸方向に移動した図5に示すようなものとなり、以下の式(5)で表される。 Here, it is assumed that the pedestrian walks at a constant speed at a speed V p (V xp , V yp ) and has only a speed component in the x direction. In this case, the distribution of the position X pt (x pt , y pt ) of the pedestrian at time t is as shown in FIG. 5 in which X po is moved in the X-axis direction, and is expressed by the following equation (5). The

ここで、時刻tにおいて自車両幅wの範囲に歩行者が存在する図6に示すような場合に衝突と判定すると、各将来時刻tにおける衝突確率Pcol(t)は、以下の式(6)により求めることができる。ここで、f(xpt)は、以下の式(7)に示すような、時刻tでの自車両位置のy軸値(yet=ye0+vye・t)における歩行者位置xptの分布であり、f(xpt)はXptの関数となる。 Here, when it is determined that a collision occurs in a case where a pedestrian exists in the range of the own vehicle width w at time t as shown in FIG. 6, the collision probability P col (t) at each future time t is expressed by the following equation (6) ). Here, f (x pt ) is the pedestrian position x pt in the y-axis value (y et = y e0 + v ye · t) of the host vehicle position at time t as shown in the following equation (7). Distribution, f (x pt ) is a function of X pt .

式(6)により、各将来時刻tにおける衝突確率が求まり、式(6)を有限の時間長Tmaxに亘って積分した以下の式(8)が、時刻t=0からt=Tmaxまでの衝突確率P(C=1)となる。 The collision probability at each future time t is obtained from the equation (6), and the following equation (8) obtained by integrating the equation (6) over a finite time length T max is obtained from the time t = 0 to t = T max. Collision probability P (C = 1).

つぎに、衝突時間演算部1bは、式(6)で得られた各将来時刻tにおける衝突確率Pcol(t)と式(8)で得られた衝突確率P(C=1)とから、衝突する場合におけるtTTCの確率分布p(tTTC|C=1)を、以下の式(9)で求め、式(9)と各衝突時刻t(当該時刻tは衝突時間tTTCそのもの)から、衝突する場合におけるtTTCの期待値を、以下の式(10)で求める(ステップSA2)。 Next, the collision time calculation unit 1b uses the collision probability P col (t) at each future time t obtained by Expression (6) and the collision probability P (C = 1) obtained by Expression (8), The probability distribution p (t TTC | C = 1) of t TTC in the case of a collision is obtained by the following equation (9), and from the equation (9) and each collision time t (the time t is the collision time t TTC itself). The expected value of t TTC in the case of a collision is obtained by the following equation (10) (step SA2).

つぎに、仮想時間設定部1cは、仮想時間を、ある固定値Tcに設定する(ステップSA3)。なお、当該固定値Tcは、車両の走行速度に応じて変化させてもよい。   Next, the virtual time setting unit 1c sets the virtual time to a certain fixed value Tc (step SA3). The fixed value Tc may be changed according to the traveling speed of the vehicle.

つぎに、衝突しない場合のtTTCの確率分布p(tTTC|C=0)を仮想時間Tcでのインパルスとして定義すると、仮想時間演算部1dは、衝突しない場合におけるtTTCの期待値を、以下の式(11)で求める(ステップSA4)。ここで、衝突しない場合に有限な時間長の衝突時間を設定することが、本実施形態の特長である。 Then, the probability distribution of t TTC when not collide p | Defining a (t TTC C = 0) as an impulse in the virtual time Tc, the virtual time computation unit 1d is the expected value of t TTC when not collide, It calculates | requires with the following formula | equation (11) (step SA4). Here, it is a feature of this embodiment that a collision time having a finite time length is set when no collision occurs.

つぎに、重みα設定部1eは、衝突する場合のtTTCの期待値に対する重みαを、以下の式(12)に示す通り、式(8)で得られた衝突確率P(C=1)に設定する(ステップSA5)。また、重みβ設定部1fは、衝突しない場合のtTTCの期待値に対する重みβを、以下の式(13)に示す通り、式(8)で得られた衝突確率P(C=1)を用いて衝突確率P(C=0)に設定する(ステップSA5)。
α=P(C=1) ・・・(12)
β=P(C=0)=1−P(C=1) ・・・(13)
Next, the weight α setting unit 1e sets the weight α for the expected value of t TTC in the case of a collision to the collision probability P (C = 1) obtained by the equation (8) as shown in the following equation (12). (Step SA5). Further, the weight β setting unit 1f uses the collision probability P (C = 1) obtained by the equation (8) as shown in the following equation (13) as the weight β for the expected value of t TTC when there is no collision. To set the collision probability P (C = 0) (step SA5).
α = P (C = 1) (12)
β = P (C = 0) = 1−P (C = 1) (13)

そして、加算部1gは、ステップSA2で演算した衝突する場合におけるtTTCの期待値E[tTTC|C=1](式(10))、ステップSA4で演算した衝突しない場合におけるtTTCの期待値E[tTTC|C=0](=Tc)(式(11))、およびステップSA5で設定した重みαとβを用いて、以下の式(14)に従って、最終的な衝突時間TTCを演算する(ステップSA6)。
TTC=α・E[tTTC|C=1]+β・E[tTTC|C=0]
=α・E[tTTC|C=1]+β・Tc ・・・(14)
Then, the addition section 1g is, t TTC expected value E in a case of collision calculated in step SA2 | expectations t TTC when [t TTC C = 1] (the formula (10)), does not collide calculated in step SA4 Using the value E [t TTC | C = 0] (= Tc) (equation (11)) and the weights α and β set in step SA5, the final collision time TTC is calculated according to the following equation (14). Calculation is performed (step SA6).
TTC = α · E [t TTC | C = 1] + β · E [t TTC | C = 0]
= Α · E [t TTC | C = 1] + β · Tc (14)

[4.本実施形態のまとめ]
以上説明したように、本実施形態では、死角からの歩行者飛び出しのような予測衝突場面(衝突が確率的に表現される場面)に対しても衝突時間TTCを求めるために、衝突予測により、衝突が予測される場合のTTCの期待値(確率平均)と衝突が予測されない場合のTTCの期待値(事前に設定した有限な時間長)との重み付き線形和を、当該予測衝突場面に対する衝突時間として求める。これにより、実際に衝突対象がないような当該場面であっても、衝突の可能性を含めた衝突時間TTCの算出が可能となる。
[4. Summary of this embodiment]
As described above, in the present embodiment, in order to obtain the collision time TTC even for a predicted collision scene such as a pedestrian jumping out from a blind spot (a scene in which a collision is expressed stochastically), by collision prediction, A weighted linear sum of an expected value of TTC (probability average) when a collision is predicted and an expected value of TTC (a finite time length set in advance) when no collision is predicted is used as a collision for the predicted collision scene. Ask as time. This makes it possible to calculate the collision time TTC including the possibility of collision even in the scene where there is no actual collision target.

ここで、論文「若林拓史ら,“交通流ビデオ解析システムを用いた交通コンフリクト分析と新しい危険度評価指標の提案”,土木計画学研究・論文集,巻:20号,4頁,pp.949−956,2003」では、先行車追従場面において高速度で車間距離が短い場面(危険場面)であっても、先行車との相対速度が極めて小さい値の場合には、衝突時間TTCは大きな値をとるため危険度評価指標として不適であるので、当該小さい値の場合には、先行車にある減速度を設定したときの衝突時間を新たな危険度評価指標とする提案がなされている。しかし、当該論文で提案された危険度評価指標は、先行車追従場面に対してのものである。また、当該論文においては、衝突することが前提(確率1で衝突する前提)であり、衝突を確率的なものとして扱っていない。さらに、当該論文では、ある確率で起こり得る衝突しない場合に対する衝突時間の扱いについては未定義である。そのため、当該論文で提案された新たな危険度評価指標を、死角からの飛び出しのような衝突対象が顕在化していない場面(予測される危険場面)に対しての危険度評価指標として用いることはできない。   Here, the paper “Takushi Wakabayashi et al.,“ Traffic Conflict Analysis Using Traffic Flow Video Analysis System and Proposal of New Risk Evaluation Index ”, Civil Engineering Planning Research Papers, Volume: 20, No. 4, p. 949 -956, 2003 ", the collision time TTC is a large value when the relative speed with the preceding vehicle is a very small value even when the preceding vehicle following scene is a high speed and the distance between the vehicles is short (dangerous scene). Therefore, in the case of the small value, there is a proposal that the collision time when the deceleration in the preceding vehicle is set is used as a new risk evaluation index. However, the risk evaluation index proposed in this paper is for the preceding vehicle following scene. Further, in this paper, the collision is a premise (the premise of collision with probability 1), and the collision is not treated as probabilistic. Furthermore, in this paper, the handling of the collision time for the case of no collision that can occur with a certain probability is undefined. Therefore, the new risk evaluation index proposed in this paper can be used as a risk evaluation index for scenes where a collision target such as popping out from a blind spot is not obvious (predicted danger scene). Can not.

そこで、本実施形態では、上述したアルゴリズムで、死角からの歩行者飛び出しのような予測衝突場面に対して衝突時間TTCを求めた。特に、この衝突時間TTCの演算の際に、衝突が予測されない場合のTTCの期待値を有限な時間長に設定する(通常では、衝突しない場合の衝突時間TTCは無限大となる。)ことで、実際に衝突対象がないような当該場面であっても、衝突の可能性を含めた衝突時間TTCの算出を可能とした。   Therefore, in the present embodiment, the collision time TTC is obtained for a predicted collision scene such as a pedestrian jumping from a blind spot with the algorithm described above. In particular, when calculating the collision time TTC, the expected TTC value when no collision is predicted is set to a finite time length (normally, the collision time TTC when there is no collision is infinite). Even in such a scene where there is no actual collision target, the collision time TTC including the possibility of the collision can be calculated.

以上のように、本発明にかかる衝突時間演算装置は、自動車製造産業において有用であり、特に、衝突時間TTCの演算を行うための利用に適している。   As described above, the collision time calculation device according to the present invention is useful in the automobile manufacturing industry, and is particularly suitable for use for calculating the collision time TTC.

1 衝突時間演算装置
1a 衝突予測部
1b 衝突時間演算部
1c 仮想時間設定部
1d 仮想衝突時間演算部
1e 重みα設定部
1f 重みβ設定部
1g 加算部
DESCRIPTION OF SYMBOLS 1 Collision time calculation apparatus 1a Collision prediction part 1b Collision time calculation part 1c Virtual time setting part 1d Virtual collision time calculation part 1e Weight alpha setting part 1f Weight beta setting part 1g Adder

Claims (1)

衝突する場合と衝突しない場合とを含めて将来の衝突時間の分布を仮定して将来の衝突時間の期待値を求め、求めた期待値を用いて予測衝突場面に対する衝突時間を演算する衝突時間演算装置において、
衝突しない場合の衝突時間を固定値に設定すること、
を特徴とする衝突時間演算装置。
Collision time calculation that calculates the collision time for the predicted collision scene using the expected value obtained by calculating the expected value of the future collision time assuming the distribution of the future collision time including the case of collision and the case of no collision In the device
Set the collision time to a fixed value when there is no collision,
A collision time calculation device characterized by the above.
JP2010088094A 2010-04-06 2010-04-06 Collision time calculation device Active JP5607410B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2010088094A JP5607410B2 (en) 2010-04-06 2010-04-06 Collision time calculation device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2010088094A JP5607410B2 (en) 2010-04-06 2010-04-06 Collision time calculation device

Publications (2)

Publication Number Publication Date
JP2011221668A true JP2011221668A (en) 2011-11-04
JP5607410B2 JP5607410B2 (en) 2014-10-15

Family

ID=45038607

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2010088094A Active JP5607410B2 (en) 2010-04-06 2010-04-06 Collision time calculation device

Country Status (1)

Country Link
JP (1) JP5607410B2 (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101866014B1 (en) 2016-03-22 2018-06-08 현대자동차주식회사 Apparatus for avoiding side crash in vehicle and method thereof

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2006070865A1 (en) * 2004-12-28 2006-07-06 Kabushiki Kaisha Toyota Chuo Kenkyusho Vehicle motion control device
JP2007257338A (en) * 2006-03-23 2007-10-04 Toyota Central Res & Dev Lab Inc Potential risk estimation device
JP2008021269A (en) * 2006-07-14 2008-01-31 Denso It Laboratory Inc Collision risk determining device, collision risk determining method and pedestrian identifying method
JP2008171207A (en) * 2007-01-11 2008-07-24 Fuji Heavy Ind Ltd Vehicle driving support device

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2006070865A1 (en) * 2004-12-28 2006-07-06 Kabushiki Kaisha Toyota Chuo Kenkyusho Vehicle motion control device
JP2007257338A (en) * 2006-03-23 2007-10-04 Toyota Central Res & Dev Lab Inc Potential risk estimation device
JP2008021269A (en) * 2006-07-14 2008-01-31 Denso It Laboratory Inc Collision risk determining device, collision risk determining method and pedestrian identifying method
JP2008171207A (en) * 2007-01-11 2008-07-24 Fuji Heavy Ind Ltd Vehicle driving support device

Also Published As

Publication number Publication date
JP5607410B2 (en) 2014-10-15

Similar Documents

Publication Publication Date Title
JP6622148B2 (en) Ambient environment recognition device
JP4416020B2 (en) Travel plan generator
US9552733B2 (en) Course evaluation apparatus and course evaluation method
JP6203381B2 (en) Route prediction device
Tamke et al. A flexible method for criticality assessment in driver assistance systems
US10829113B2 (en) Vehicle collision avoidance
JP5212753B2 (en) Radar apparatus and measurement method used in the radar apparatus
WO2015156097A1 (en) Collision prevention device
JP5146542B2 (en) Traveling route estimation device and traveling route estimation method used in the device
US10351129B2 (en) Collision mitigation and avoidance
JP5402813B2 (en) Vehicle motion control device and program
JP2011227582A (en) Vehicle collision prediction apparatus
KR101793827B1 (en) Method for assessing collision risk and avoiding collision based on roadway geometry
CN113096424A (en) Automatic emergency braking method and system for pedestrian crossing vehicle
WO2019171097A1 (en) Vehicle behavior prediction method and vehicle behavior prediction device
JP6477253B2 (en) Vehicle travel control device
JP2021030956A (en) Operation support apparatus
JP5607410B2 (en) Collision time calculation device
WO2018198769A1 (en) Surrounding environment recognition device, display control device
CN115447607A (en) Method and device for planning a vehicle driving trajectory
JP6306429B2 (en) Steering control system
JP2018005390A (en) Object tracking method and object tracking apparatus
JP2022052875A (en) Automated driving apparatus
JPWO2019186716A1 (en) In-vehicle device, information processing method, and information processing program
JP2010224786A (en) Collision risk degree estimating device and program

Legal Events

Date Code Title Description
A621 Written request for application examination

Free format text: JAPANESE INTERMEDIATE CODE: A621

Effective date: 20130207

A131 Notification of reasons for refusal

Free format text: JAPANESE INTERMEDIATE CODE: A131

Effective date: 20131224

A977 Report on retrieval

Free format text: JAPANESE INTERMEDIATE CODE: A971007

Effective date: 20131226

A521 Request for written amendment filed

Free format text: JAPANESE INTERMEDIATE CODE: A523

Effective date: 20140212

A131 Notification of reasons for refusal

Free format text: JAPANESE INTERMEDIATE CODE: A131

Effective date: 20140415

A521 Request for written amendment filed

Free format text: JAPANESE INTERMEDIATE CODE: A523

Effective date: 20140530

TRDD Decision of grant or rejection written
A01 Written decision to grant a patent or to grant a registration (utility model)

Free format text: JAPANESE INTERMEDIATE CODE: A01

Effective date: 20140805

A61 First payment of annual fees (during grant procedure)

Free format text: JAPANESE INTERMEDIATE CODE: A61

Effective date: 20140828

R151 Written notification of patent or utility model registration

Ref document number: 5607410

Country of ref document: JP

Free format text: JAPANESE INTERMEDIATE CODE: R151

R250 Receipt of annual fees

Free format text: JAPANESE INTERMEDIATE CODE: R250

R250 Receipt of annual fees

Free format text: JAPANESE INTERMEDIATE CODE: R250

R250 Receipt of annual fees

Free format text: JAPANESE INTERMEDIATE CODE: R250

R250 Receipt of annual fees

Free format text: JAPANESE INTERMEDIATE CODE: R250

R250 Receipt of annual fees

Free format text: JAPANESE INTERMEDIATE CODE: R250

R250 Receipt of annual fees

Free format text: JAPANESE INTERMEDIATE CODE: R250

R250 Receipt of annual fees

Free format text: JAPANESE INTERMEDIATE CODE: R250