DE102017118651A1 - Method and system for collision avoidance of a vehicle - Google Patents

Abstract

The present invention relates to a method for collision avoidance of a vehicle (10), the method comprising the steps of: measuring at least one parameter describing the dynamic states of the vehicle (10) and predicting a vehicle trajectory (22) in response to the dynamic States of the vehicle (10); Detecting at least one object (24) in an environmental region (26) of the vehicle (10) and determining at least one parameter describing the dynamic states of the detected object (24) relative to the vehicle (10), wherein the environmental region (26) of the vehicle (10) includes at least one peripheral region (28) remote from the predicted vehicle trajectory (22); Predicting a potential collision between the vehicle (10) and the detected object (24) based on the dynamic conditions of the vehicle (10) and the dynamic conditions of the object (24); and causing at least one collision avoidance countermeasure in response to the predicted potential collision. The prediction of a possible collision with an object (24) detected in the peripheral area (28) includes determining the location (32) of the possible collision on the vehicle trajectory (22), the countermeasure for avoiding a collision with the one in the peripheral area (FIG. 28) in response to a suspected potential collision with such object (24) at the particular location (32).
The present invention also relates to a corresponding computer program product for carrying out the method and to a corresponding system (34) for collision avoidance.

Description

The present invention relates to a method for collision avoidance of a vehicle. The method comprises the steps of: (i) measuring at least one parameter describing the dynamic states of the vehicle and predicting a vehicle trajectory in response to the dynamic states of the vehicle; (ii) detecting at least one object in an environmental region of the vehicle and determining at least one parameter describing the dynamic states of the detected object relative to the vehicle, the environmental region of the vehicle including at least one peripheral region remote from the predicted vehicle trajectory; (iii) predicting a potential collision between the vehicle and the detected object based on the dynamic conditions of the vehicle as well as the dynamic states of the object; and (iv) causing at least one collision avoidance countermeasure in response to the predicted potential collision.

The present invention further relates to a corresponding computer program product and a corresponding system for collision avoidance of a vehicle.

The document WO 2006/053652 A1 describes a method of avoiding collisions or reducing the amount of vehicle collision, comprising the steps of: detecting the speed and direction of travel of the vehicle and determining a vehicle trajectory in the near future; Detecting the position of objects in the environment of the vehicle; Detecting the speed and direction of movement of the objects relative to the vehicle; Predicting the future position of the objects relative to the vehicle; Evaluating the current and future positions of the objects relative to the vehicle; Issuing a warning to the driver and / or performing an automatic steering and / or braking intervention by means of a vehicle assistance system according to the evaluation, if it was determined according to the evaluation that a collision with the object without intervention by the system is unavoidable. The prediction of the future position of the objects relative to the vehicle is based on a recursive filter, such as a Kalman filter. The document also describes a vehicle assistance system for avoiding collisions or reducing the extent of vehicle collision.

It is an object of the present invention to provide a method and system for easy and safe collision avoidance.

This object is solved by the independent claims. Advantageous embodiments are given in the dependent claims.

According to various aspects of the method of collision avoidance of a vehicle according to the invention, the method comprises the steps of: (i) measuring at least one parameter describing the dynamic states of the vehicle and predicting a vehicle trajectory in response to the vehicle's dynamic states; (ii) detecting at least one object in an environmental region of the vehicle and determining at least one parameter describing the dynamic states of the detected object relative to the vehicle, wherein the environmental region of the vehicle includes at least one peripheral region remote from the predicted vehicle trajectory, (iii) predicting a possible collision between the vehicle and the detected object based on the dynamic states of the vehicle and the dynamic states of the object; and (iv) causing at least one collision avoidance countermeasure in response to the predicted potential collision. The prediction of a possible collision with an object detected in the peripheral area remote from the predicted vehicle trajectory involves determining the position of the potential collision on the vehicle trajectory. The countermeasure for avoiding a collision with such an object detected in the peripheral area remote from the predicted vehicle trajectory is made in response to a suspected potential collision at the designated location. Typical parameters for describing the dynamic states are speed, acceleration, etc.

The basic idea of the invention is to provide a method for collision avoidance, which offers the corresponding system sufficient time to initiate the necessary countermeasure for collision avoidance, in particular to predict and set the speed of the vehicle. This is realized by the above-mentioned estimation of the location of a possible collision with objects from a peripheral area of the surrounding area "in advance" of the vehicle.

According to a preferred embodiment of the invention, the method comprises the further step of predicting an object trajectory taking into account the dynamic states of the object, wherein the location of the possible collision on the vehicle trajectory is the intersection between the vehicle trajectory and the object trajectory. In other words, the location of the possible collision is given by a virtual representation of the object, which is located at the intersection between the vehicle trajectory and the object trajectory.

According to another preferred embodiment of the invention, the object or at least one of the objects is detected as another vehicle. This vehicle may be a car, a motorcycle, a truck, etc.

According to a particularly preferred embodiment of the invention, the method comprises the further steps of detecting a maneuver intent of the other vehicle and the corresponding maneuver type and taking into account the maneuver intent and the corresponding maneuver type in the prediction of the object trajectory of the other vehicle. Typical maneuvers are lane change, crossing the lane / road, driving onto the lane, etc.

According to another preferred embodiment of the invention, the peripheral area removed from the predicted vehicle trajectory of the vehicle traveling on a road is another lane of the road or a shoulder of the road or another road.

According to a further preferred embodiment of the invention, the method further comprises the further step of selecting the chronological sequence of possible collisions with objects which have to be taken into account by means of countermeasures. By the mentioned estimation of the location of a possible collision with objects from a peripheral area, it is possible to estimate whether an object and which specific object is the most critical object in terms of the risk of a collision.

According to another preferred embodiment of the invention, the location of the possible collision is determined taking into account the outline of the vehicle and / or the outline of the object. This is especially important for large objects and small angles between the two corresponding trajectories.

In accordance with yet another preferred embodiment of the invention, the at least one object is detected using a radar system, a lidar system, a camera system and / or other suitable sensor system. In this connection, all these systems are used as surveying systems for measuring the distance to a destination.

The collision avoidance countermeasure preferably includes at least one of the following: emitting a warning message, performing an automatic brake maneuver, and performing an automatic steering maneuver.

The method is preferably a method for collision avoidance during an autonomous driving operation of the vehicle.

The computer program product of the invention comprises computer executable program code portions having program code instructions configured to execute the aforementioned method.

The present invention further relates to a system for collision avoidance of a vehicle, the system being configured to perform the steps of: (i) measuring at least one parameter describing the dynamic conditions of the vehicle and predicting a vehicle trajectory in response to the dynamic conditions of the vehicle; (ii) detecting at least one object in an environmental region of the vehicle and determining at least one parameter describing the dynamic states of the detected object relative to the vehicle, the environmental region of the vehicle including at least one peripheral region remote from the predicted vehicle trajectory; (iii) predicting a potential collision between the vehicle and the detected object; and (iv) initiating countermeasures for collision avoidance in response to the predicted potential collision. The prediction of a potential collision with an object detected in the peripheral area remote from the predicted vehicle trajectory involves determining the position of the potential collision on the vehicle trajectory, the countermeasure for avoiding collision with such an object located in the vehicle body the predicted vehicle trajectory remote peripheral region is detected, in response to a suspected possible collision with such an object is made at the specific location.

The system is preferably a system for collision avoidance during autonomous driving of the vehicle.

Further features of the invention will become apparent from the claims, the figures and the description of the figures. All of the features and feature combinations mentioned above in the description as well as the features and feature combinations mentioned below alone in the description of the figures and / or in the figures are usable not only in the respectively specified combination but also in other combinations or alone.

• 1 eine Draufsicht auf eine Straßenszene mit zwei Fahrzeugen, wobei eines der Fahrzeuge ein System zur Kollisionsvermeidung gemäß einer bevorzugten Ausführungsform der Erfindung aufweist;
• 2 eine Draufsicht zum Darstellen eines Algorithmus für eine Trajektorienvorhersage mit den beiden in 1 dargestellten Fahrzeugen; und
• 3 ein Blockdiagramm eines Systems zur Kollisionsvermeidung gemäß einer bevorzugten Ausführungsform der Erfindung.

The invention will now be explained in more detail with reference to a preferred embodiment and with reference to the accompanying drawings. Show it:

• 1 a plan view of a street scene with two vehicles, one of the vehicles having a system for collision avoidance according to a preferred embodiment of the invention;
• 2 a plan view illustrating an algorithm for a trajectory prediction with the two in 1 illustrated vehicles; and
• 3 a block diagram of a system for collision avoidance according to a preferred embodiment of the invention.

1 shows a plan view of a vehicle 10 that on a street 12 moves. This vehicle 10 is a motor vehicle, more specifically a car. The road 12 is a multi-lane road and has two lanes 14 . 16 , a first lane 14 by the vehicle 10 is used, and a second lane 16 for traffic in the same direction. At the edge of every lane 14 . 16 there is a side strip 18 the street 12 , On the verges 18 next to the lane 14 by the vehicle 10 is used, there is another vehicle 20 , The vehicle 10 drives on the first lane 14 along a path whose further course through a predicted vehicle trajectory 22 given is. The vehicle has several sensors for detecting vehicles in the area. The vehicle 10 has a sensor system (not shown) for detecting objects 24 (like the other vehicle 20 ) in a surrounding area 26 in the advance direction of the vehicle 10 on. The environment area 26 has a central area along a portion of the predicted vehicle trajectory 22 and two peripheral areas 28 left and right of the predicted vehicle trajectory 22 on. The other vehicle 20 travels along a path whose further course through a predicted object trajectory 28 given is.

The vehicle 10 has a system (in the 1 and 2 not shown) to avoid collision with detected objects 24 , especially with other vehicles 10 , on. The system is configured to perform the following four steps: (i) measuring at least one parameter that is representative of the dynamic conditions of the vehicle 10 describes and predicting the vehicle trajectory 22 in response to the dynamic conditions of the vehicle 10 ; (ii) Capture objects 24 , such as the other vehicle 20 in the surrounding area 26 in advance, and determining at least one parameter representing the dynamic states of the detected object (s) 24 relative to the vehicle 10 describes; (iii) predicting a possible collision between the vehicle 10 and the detected object 24 , wherein the prediction of a possible collision with an object 24 that in the predicted vehicle trajectory 22 distant peripheral area 28 is detected (such as the other vehicle 20 that on the side strip 18 the street 12 detecting) determining the location 32 the possible collision on the vehicle trajectory 22 includes; and (iv) initiating countermeasures for collision avoidance in response to the predicted potential collision, the countermeasure for avoiding collision with such object 24 that in the peripheral area 28 in response to a predicted potential collision with such an object at the particular location 32 is initiated. Typical countermeasures for collision avoidance may be automatic braking maneuvers and / or automatic steering maneuvers.

The place 32 the possible collision on the vehicle trajectory 22 is determined by predicting an object trajectory 30 taking into account the dynamic states of the object 24 determined, where the place 32 the possible collision on the vehicle trajectory 22 the intersection between the vehicle trajectory 22 and the object trajectory 30 is. The place 32 the possible collision is taking into account the outline of the vehicle 10 and the outline of the object 24 certainly. In the 1 shown collision prediction uses a trajectory prediction according to the "least squares method".

The sensor system (not shown) for detecting objects 24 (such as the other vehicle 20 ) in the surrounding area 26 in advance of the vehicle 10 is a radar system (radar: Radio Detection and Ranging), a lidar system (Lidar: Light Detection and Ranging) or a camera system, such as a 3D camera system.

2 shows a top view of a fairly similar street scene with the two in 1 illustrated vehicles 10 . 20 , The difference to 1 is the predicted object trajectory 30 , which correspond to the maneuver intent of the other vehicle 20 arcuate, namely a change from the side strip 18 on the by the vehicle 10 used lane 14 , The system for collision avoidance of the vehicle 10 detects this maneuver intention of the other vehicle 20 and the corresponding maneuvering type from the dynamic states of the other vehicle 20 and considers the maneuver intention and the corresponding maneuver type in the prediction of the object trajectory 24 of the other vehicle 20 , That is, the system says another object trajectory 30 of the other vehicle 20 before, with the result that the intersection of the vehicle trajectory 22 and the object trajectory 30 in relation to the place 32 the possible collision on the in 1 certain vehicle trajectory is offset.

3 shows a high-level block diagram of a system 34 for the collision avoidance of a vehicle 10 with objects 24 ,

The collision prediction system 34 based on the following modules: a detection module 36 to capture the objects 24 , a trajectory prediction module 38 to perform a trajectory prediction of the objects 24 and a trajectory prediction of the ego vehicle 10 , This system 34 is for a classical trajectory prediction (as in 1 as well as being suitable for trajectory prediction involving maneuvering intentions (as in 2 is shown).

A collision prediction module 40 for predicting a collision between the vehicle 10 and the object (objects) 24 uses an algorithm that calculates the probability of shape collisions using the data (same position, time, velocity, etc.) given by the trajectory predictions.

The shape collision algorithm integrates the safety distance around the ego vehicle 10 (front, side and rear distance). A corresponding collision detection module 42 calculated for each sampling time of the trajectory predictions (prediction of the vehicle trajectory 22 and predicting the object trajectories 30 ), whether the projected shape of the vehicle 10 / objects 24 collide or not. If no collision is predicted, the system becomes 36 proceed to the next detected object (path N).

If a collision is predicted (path Y), a corresponding module is used 44 a projection of a virtual object 24 ' generates: the object 24 or more than one object 24 the object with the highest probability of collision, becomes the vehicle trajectory 22 projected. This object 24 is sent to a destination selector module 46 which determines the object of interest (the object 24 in terms of the control by a control module 48 is performed). This allows the control module 48 to control the speed, or a maneuver planning module (not shown) to avoid the collision, eg by a steering maneuver. All mentioned modules are preferably realized as software modules.

The group 50 of modules 38 . 40 . 42 . 44 becomes for objects 24 used only in a peripheral area 28 be recorded. objects 24 located in the central area along the vehicle trajectory 22 are always objects 24 with high collision probability. These objects 24 are collected directly from the acquisition module after their acquisition 36 to the destination selection module 46 transfer.

• (a) „Geschwindigkeitsregelung“ zum Steuern der Geschwindigkeit des Fahrzeugs 10 bei einer vom Fahrer gewählten Geschwindigkeit, sie wird hauptsächlich verwendet, falls die Straße 12 frei ist, und
• (b) „Abstandsregelung“ zum Halten eines ausreichenden Abstands zwischen dem Ego-Fahrzeug 10 und einem Ziel/Objekt 24 auf der gleichen Fahrspur 14 (z.B. durch Sicherstellen eines ausreichenden Zeitabstands zwischen Fahrzeugen).

That is, the group 50 of modules 38 . 40 . 42 . 44 Can be considered an upgrade to an existing system with a collection module 36 , a destination acquisition module 46 and a control module 48 be understood. The control module 48 Such an existing system is preferably a typical linear regulator with two main functions and corresponding modes:

• (a) "cruise control" for controlling the speed of the vehicle 10 at a speed selected by the driver, it is mainly used if the road 12 is free, and
• (b) "distance control" for keeping a sufficient distance between the ego vehicle 10 and a target / object 24 on the same lane 14 (eg by ensuring sufficient time between vehicles).

The destination selection module 46 is a the longitudinal control module 48 upstream component. The destination selection module 46 chooses among the numerous recorded vehicles 20 or other detected objects 24 out, which by the longitudinal control module 48 must be taken into account. If the road 12 is free, the longitudinal control module 48 set to the speed control mode and therefore does not consider information about detected objects 24 (sometimes referred to as destination information).

The destination selection module 46 that the objects 24 for the classic distance control of a longitudinal control module 48 is mainly about information about objects 24 on the vehicle trajectory 22 or the lane 14 does not care about and uses any information about this trajectory 22 removed objects 24 (eg objects 24 on side tracks, side strips, etc.).

The task of additional collision detection (Modul 42 ) is these objects 24 to take into account, as they also have an impact on the ego vehicle 10 can have, for example if the other vehicle 20 on the verges 18 or on a side lane with a lane change begins.

The main function of the additional modules 38 to 44 (Basic family 50 ) is therefore the system 34 to allow the risk of collision by regulating the speed of the vehicle 10 to reduce in advance. This is what the trajectory prediction module says 38 the object trajectory 30 every object 24 ahead, and then verifies the collision prediction module 40 whether one of these trajectories 30 the (ego) vehicle trajectory 22 will cross.

If a possible collision by the module 40 is captured, the system generates 34 a virtual object 24 ' That's on the vehicle trajectory 22 is arranged (by using the module 44 ). Thus, this "target" becomes the target selection module 46 and then from the longitudinal control 48 considered if this virtual object 24 ' is considered relevant. The virtual object 24 ' has all the characteristics of the corresponding object 24 , except the position of the object 24 ,

LIST OF REFERENCE NUMBERS

vehicle 10 Street 12 lane 14 lane 16 shoulder 18 another vehicle 20 vehicle trajectory 22 object 24 virtual object 24 ' surrounding area 26 peripheral area 28 object trajectory 30 Place of a possible collision 32 Collision prediction system 34 acquisition module 36 Trajektorienvorhersagemodul 38 Collision prediction module 40 Collision detection module 42 Module for creating virtual objects 44 Target selection module 46 control module 48 module group 50 No N Yes Y

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• WO 2006/053652 A1 [0003]

Claims (12)

A method of collision avoidance of a vehicle (10), the method comprising the steps of: measuring at least one parameter describing the dynamic states of the vehicle (10) and predicting a vehicle trajectory (22) in response to the vehicle's dynamic states ( 10); Detecting at least one object (24) in a surrounding area (26) of the vehicle (10) and determining at least one parameter describing the dynamic states of the detected object (24) relative to the vehicle (10), wherein the surrounding area (26) of the Vehicle (10) includes at least one peripheral region (28) remote from the predicted vehicle trajectory (22); - predicting a possible collision between the vehicle (10) and the detected object (24) based on the dynamic states of the vehicle (10) and the dynamic states of the object (24); and - causing at least one collision avoidance countermeasure in response to the predicted potential collision; characterized in that the prediction of a possible collision with an object (24) detected in the peripheral area (28) comprises determining the location (32) of the potential collision on the vehicle trajectory (22), the countermeasure for avoiding a collision with such in the peripheral region (28), the object (24) is caused to respond to a suspected potential collision with such an object (24) at the particular location (32). Method according to Claim 1 characterized by the further step of predicting an object trajectory (30) taking into account the dynamic states of the object (24), wherein the location (32) of the possible collision on the vehicle trajectory (22) is the intersection between the vehicle trajectory (22) and the object trajectory (30). Method according to Claim 1 or 2 , characterized in that the object (24) or at least one of the objects is detected as another vehicle (20). Method according to Claim 3 characterized by the further steps of: detecting a maneuver intent of the other vehicle (20) and the corresponding maneuver type; and - taking into account the maneuver intent and the corresponding maneuver type in the prediction of the object trajectory (24) of the other vehicle (20). Method according to one of Claims 1 to 4 characterized in that the peripheral area (28) remote from the predicted vehicle trajectory (22) of the vehicle (10) traveling on a road (12) - another lane (16) of the road (12); or - a side strip (18) of the road (12); or - covering another street. Method according to one of Claims 1 to 5 characterized by the further step of selecting the chronological order of possible collisions with objects (24) which must be taken into account by countermeasures. Method according to one of Claims 1 to 6 , characterized in that the location (32) of the potential collision is determined taking into account the outline of the vehicle (10) and / or the outline of the object (24). Method according to one of Claims 1 to 7 , characterized in that the at least one object (24) is detected using a radar system and / or a lidar system and / or a camera system. Method according to one of Claims 1 to 8th , characterized in that the countermeasure for collision avoidance includes at least one of the following measures: - transmitting a warning message, in particular to the driver of the vehicle (10); - Carry out an automatic braking maneuver; and - performing an automatic steering maneuver. Method according to one of Claims 1 to 8th characterized in that the method is a method for collision avoidance during an autonomous driving operation of the vehicle (10). A computer program product comprising computer executable program code portions having program code instructions configured to perform the method of any one of Claims 1 to 10 perform. A system (34) for collision avoidance of a vehicle (10), the system (34) configured to perform the steps of: measuring at least one parameter describing the dynamic conditions of the vehicle (10) and predicting a vehicle trajectory (22 in response to the dynamic conditions of the vehicle (10); Detecting at least one object (24) in a surrounding area (26) of the vehicle (10) and determining at least one parameter describing the dynamic states of the detected object (24) relative to the vehicle (10), wherein the surrounding area (26) of the Vehicle (10) includes at least one peripheral region (28) remote from the predicted vehicle trajectory (22); - predicting a possible collision between the vehicle (10) and the detected object (24); and - causing countermeasures to avoid collision in response to the predicted potential collision; characterized in that the prediction of a possible collision with an object (24) detected in the peripheral area (28) comprises determining the location (32) of the potential collision on the vehicle trajectory (22), the countermeasure for avoiding a collision with a vehicle peripheral area (28) detected object (24) in response to a suspected possible collision with such an object (24) at the particular location (32) is caused.

Images