DE102020210336A1 - Collision avoidance for a vehicle - Google Patents

Abstract

A method (300) for collision avoidance of a vehicle (105) comprises the steps of scanning a lateral environment (115) of the vehicle (105); detecting a first object (220) moving relative to the surroundings (115) and which is at least partially covered by a second object (230); determining a prospective trajectory (225) of the first object (220); determining an expected trajectory (210) of the vehicle (105); and determining, based on the trajectories (210, 225), whether a collision of the vehicle (105) with the first object (220) is imminent.

Description

The present invention relates to collision avoidance for a vehicle. In particular, the invention relates to the detection of an object in the area surrounding the vehicle.

A vehicle driving along a road on public roads, for example, runs the risk of colliding with another vehicle approaching it from the side if visibility between the vehicles is restricted. In urban areas, for example, an information sign or a fence in the area of a street crossing can make it difficult to detect the other vehicle at an early stage. In rural areas, trees, bushes, animals or other objects can be in the field of vision between the vehicles. A driver of the vehicle may be late in noticing the other vehicle, so that he cannot control his own vehicle to surely avoid a collision.

One of the objects on which the present invention is based is the improved avoidance of a collision between a vehicle and an object that is moving in its vicinity. The invention solves this problem by means of the subject matter of the independent claims. Subclaims reflect preferred embodiments.

According to a first aspect of the present invention, a method for avoiding a collision of a vehicle comprises the steps of scanning a lateral environment of the vehicle; detecting a first object moving relative to the environment, which is at least partially covered by a second object; determining a prospective trajectory equation of the first object; determining an anticipated trajectory of the vehicle; and determining, based on the trajectories, whether the vehicle is about to collide with the first object.

It was recognized that the vehicle can be better protected against a particularly lateral impact on the first object by detecting the first object and predicting its movement. The detection and determination of the first object can take place in particular by machine, in which case a sensitive sensor can be used which can be more sensitive than the eye of a driver, in particular in the area surrounding the vehicle. The detection of the first object or the prediction of its probable trajectory can take place automatically, for example using methods of pattern recognition, which can be implemented using an artificial neural network, for example.

The first object can be at a predetermined minimum distance from the anticipated trajectory of the vehicle, in particular in the lateral direction of the trajectory. This can secure an area of the vehicle that is not well covered by a driver's main field of vision or a forward-facing only sensor. Improved protection against a collision, in particular from the side, can thus be achieved.

In a first variant, a section of the first object is covered by the second object. For example, a top, bottom, front, or back portion of the first object may be obscured by the second object. The first object can preferably be perceived by the second object in a large number of sections, it being possible for a further large number of sections to be covered. For example, the second object can include a bush, a fence, a tree or a shrub that covers individual sections of the first object in a patterned manner. Due to the movement of the first object relative to the second object, a section that is covered at a first point in time can be detected at a second point in time. By scanning the first object covered in sections as many times as possible at different points in time, the object can nevertheless be recognized and its trajectory can be determined. Usually, the second object is stationary in relation to the environment, but it can also be moving, for example in the case of fog, smoke, falling leaves or a crowd.

In a second variant, the first object can be detected over a predetermined period, it being covered by the second object for part of the period. In other words, temporally offset samples of the first object can be provided, from which information about the first object can be combined.

In a further specific embodiment, a probability is determined on the basis of the two trajectories with which a collision of the vehicle with the first object is imminent. The probability can take into account a certainty of determination, which usually decreases with an increasing time interval from a current point in time. As the reliability of the determination decreases, the probability determined can also decrease. A quantitative indication of the risk of an impending collision can thus be determined.

In order to determine the probable trajectory of the vehicle, information available on board the vehicle in particular can be used be applied, such as a geographic position, the course of a traveled road, an acceleration or an applicable right of way regulation in the vicinity of the vehicle. The probable trajectory of the first object can be determined depending on the recorded information. In addition, it can be determined, for example, what type of road user the first object is, for example a pedestrian, a cyclist, a motor vehicle or an animal, and its probable trajectory can be determined in an improved manner on this basis. In particular, map information in the area of the first object can be used for this purpose, which can include an indication of a traffic route of the specific road user.

If the determined probability exceeds a predetermined threshold value, it can be determined that a collision is imminent. In the event of an impending collision, a signal can be issued on board the vehicle. The signal may be directed to a driver of the vehicle or to a controller for the vehicle.

In a further embodiment, a signal is provided which indicates a time interval before an imminent collision. For example, an acoustic signal can be provided, the pitch, volume or intensity of which depends on the time interval up to the specific imminent collision. For example, depending on the time interval, the pitch can rise or the volume can increase. In another embodiment, the signal is dependent on the determined likelihood of an imminent collision. These signals, which are proportional to the distance in terms of frequency and amplitude, enable the driver to intuitively assess the possible dangerous situation. Both embodiments can be combined with each other. The signal can also be given in other ways than acoustically, for example optically or haptically.

The signal can also be directed to an automatic system on board the vehicle instead of to a driver. In one embodiment, the vehicle is controlled to avoid the determined impending collision. For this purpose, it is possible in particular to intervene in a longitudinal control of the vehicle, preferably by braking the vehicle. In a further embodiment, an intervention in a transverse control is also possible, in particular by actuating a steering device.

According to a second aspect of the present invention, a device for avoiding collision of a vehicle comprises a first sensor for scanning a lateral environment of the vehicle; an output device for providing a signal; and a processing device. In this case, the processing device is set up to detect a first object that is moving relative to the surroundings and is at least partially covered by a second object; determine an expected trajectory of the first object; determine an anticipated trajectory of the vehicle; determine whether the vehicle is about to collide with the first object based on the trajectories; and in this case to provide a signal.

The processing device can be set up to carry out a method described herein in whole or in part. For this purpose, the processing device can comprise a programmable microcomputer or microcontroller and the method can be present in the form of a computer program product with program code means. The computer program product can also be stored on a computer-readable data carrier. Features or advantages of the method can be transferred to the device or vice versa.

The device can be used in particular on board a vehicle. The vehicle can in particular include a motor vehicle, for example a passenger car, a truck, a bus or a motorcycle. The first object can include another road user, in particular another vehicle. The second object is usually firmly connected to the environment, but can be movable, for example in the form of a tree or bush, which can be moved by the wind, but whose position cannot be changed.

The first sensor can emit waves and receive reflected waves, with the second object being at least partially transparent to the emitted waves. Partial transparency is when a wave is able to penetrate the second object, losing some of its energy in the process. The energy loss can be due to reflection or absorption at the second object. Part of a wave passing through the second object cannot reach the first sensor, so the energy determined by it is reduced.

A sensor that provides a measurement signal itself is also referred to as an active sensor. The waves can in particular be electromagnetic waves which can be emitted in a predetermined frequency range which can be selected in particular as a function of the first and/or second object. The frequency range is preferably selected in such a way that the waves are reflected as well as possible on the first object, and at the same time second object can pass as well as possible. If the first object is a vehicle, for example, then waves can be used in a frequency range that is well reflected by steel, glass or plastic. If the second object includes, for example, organic material, in particular plants or animals, the frequency of the waves can be tuned to penetrate this well. In the combination mentioned by way of example, waves from the X-band can be used, for example.

In a further embodiment, the first sensor is set up to emit waves of different frequencies in order to at least partially penetrate different materials. The frequency can be changed in steps or continuously. For example, dedicated waves of different frequencies can be used to penetrate a bush or a billboard. In a further development of the specified embodiment, a material of the second object can be determined on the basis of its transparency with respect to waves of different frequencies.

It is particularly preferred that the first sensor comprises a radar sensor. Radar sensors for vehicles are well known and are commonly used to detect objects lying in the direction of travel. A side lobe of a directional characteristic of the radar sensor can be used to detect a first object located to the side of the vehicle. The side lobe encloses an angle with the intended main lobe of the radar sensor, which points in the main scanning direction, ie usually in the direction of travel. Radar signals in a sidelobe direction are usually less strong than in the main direction and the radar sensor may be less sensitive to radar waves coming from a sidelobe direction.

In a further preferred embodiment, the device also includes a second sensor for scanning the lateral surroundings, the second sensor including an infrared sensor. In this case, the first object is preferably detected on the basis of scans from both sensors. In particular, the infrared sensor can comprise a forward-looking infrared (FLIR) sensor, as opposed to a side-scanning infrared sensor. The infrared sensor is preferably sensitive in the near IR range and more preferably not cryogenically cooled. For the detection of the first object, the second object can be non-transparent (opaque) for the second sensor. Sensor signals from the two sensors can be merged with one another in order to provide a particularly revealing image of the lateral surroundings of the vehicle. The first object can be recognized better on this image, especially if it moves in relation to the environment. A combination of radar sensor and infrared sensor could be called thermographic radar.

In yet another preferred embodiment, a display device is provided for providing a graphical representation of the first object. The representation can be determined on the basis of signals from the first and/or the second sensor. The display device is preferably directed at a driver of the vehicle. For example, a head-up display or a monitor mounted in the driver's peripheral field of vision can be used for this purpose. In one embodiment, a monitor is used, which is attached essentially where the driver can usually see an outside mirror. The exterior mirror can be replaced by a camera, the output of which can be displayed on the monitor. The information of the device described herein can be displayed exclusively on the monitor or overlay or supplement an image generated in some other way.

In yet another embodiment, the first object can be detected. In particular, it can be recognized what type of road user the first object is. Information about the size, shape and movement of the first object can be used for this. An indication of the identified road user can be output on the display device, for example in textual or symbolic form.

According to yet another aspect of the present invention, a vehicle includes an apparatus as described herein.

• 1 ein Fahrzeug mit einer Vorrichtung zur Kollisionsvermeidung;
• 2 ein System; und
• 3 ein Ablaufdiagramm eines Verfahrens zur Kollisionsvermeidung eines Fahrzeugs

darstellt.The invention will now be described in more detail with reference to the attached figures, in which:

• 1 a vehicle with a collision avoidance device;
• 2 a system; and
• 3 a flow chart of a method for collision avoidance of a vehicle

represents.

1 shows a system 100, which includes a vehicle 105 and a device 110 for avoiding a collision of the vehicle 105 with an object in its surroundings 115. The vehicle 105 is preferably designed as a motor vehicle and can in particular include a passenger car. Typically, the vehicle 105 is controlled by a human driver 120; if necessary, however, a driving function can also be carried out by an automatic controller.

In the embodiment shown, the device 110 comprises a processing device 125 which is connected to a first sensor 130, an optional second sensor 135 and preferably an output device 140. Output devices 140 illustrated by way of example include an acoustic output device 145 and a display device 150. The output devices 140 are usually configured to provide a signal to the driver 120 which indicates a possible or probable collision of the vehicle 105 with another object. The signal can be provided acoustically, optically and/or haptically by means of suitable output devices 140 . In a further specific embodiment, an interface is provided in order to provide a signal to a control device of the vehicle 105 . The control device can be set up to change a trajectory of the vehicle 105 in order to avoid an impending collision with an external object.

Sensors 130 and 135 are preferably set up to scan a lateral area of vehicle 105 in particular. In contrast, a region located essentially in front of or behind vehicle 105, in particular along a trajectory of vehicle 105, cannot be taken into account. The first sensor 130 preferably includes a radar sensor that is set up to emit radar waves in the area 115 and to receive reflected radar waves. A frequency of the radar waves can be predetermined or variable. Usually, radar waves are used in the frequency range between a few and several hundred gigahertz. The radar sensor 130 can be set up to use one of several predetermined frequencies or to freely control the frequency used within a predetermined range.

The second sensor 135 preferably includes an infrared sensor, which is further preferably set up to scan the surroundings 115 in the near infrared range. It is usually a passive sensor that does not illuminate the area 115 with infrared light.

It is proposed that device 110 be set up to use sensors 130, 135 to detect an at least partially concealed object moving in surroundings 115 of vehicle 105 and to determine its probable trajectory. This is to be compared with an expected trajectory of the vehicle 105 in order to determine whether there is a risk of the vehicle 105 colliding with the object. Corresponding information should be provided by means of an output device 140 so that the driver 120 or a control device on board the vehicle 105 intervenes appropriately in order to avert the risk of a collision.

2 Figure 1 shows the system 100 in an example situation. The vehicle 105 is on a first road 205 and is driving along a probable trajectory 210 that follows the first road 205, for example. A first object 220, which in the present case is formed by another vehicle, is moving on a second road 215, which leads in from the right in a T-shape. The other vehicle 220 follows an expected trajectory 225, which can lead, for example, along the second road 215. A second object 230 is located on a direct connection between the vehicle 105 and the other vehicle 220, which in the present case is embodied as vegetation, in particular as a tree or bush. For the driver 120 of the vehicle 105, the additional vehicle 220 can only be recognized with difficulty or not at all because the second object 230 is obstructed. If both vehicles 105, 220 follow their probable trajectories 210, 225, then a collision at a collision location 235 drawn in as an example is to be feared.

Device 110 scans surroundings 115 of vehicle 105 using first sensor 130, which emits radar waves in different, predetermined directions and receives reflected radar waves. Shown symbolically in 2 are several radar lobes 240, which show the directions and transmission and reception strengths of the radar waves by way of example. A main lobe is directed in the direction of travel of the vehicle 105, side lobes are directed more laterally.

The second object 230 can be less opaque or not at all opaque for the radar signals used, so that the radar waves can penetrate the second object 230 . Samples of the first sensor 130 can be supported by those of the second sensor 135 . The sensor signals can be merged with one another in order to be able to distinguish the first object 220 from the surroundings 115 as clearly as possible.

3 shows a flowchart of a method 300 for avoiding a collision of a vehicle 105 with an external object 220. The method 300 can be carried out in particular by means of a device 110 in a system 100.

In a step 305 the surroundings 115 of the vehicle 105 can be scanned by means of the first sensor 130 . The first sensor 130 can be driven in different directions, with different intensities and using radar waves of different frequencies to work. Respectively received, reflected radar waves can be received and evaluated.

In addition to this, in a step 310 the surroundings 115 can be scanned by means of the second sensor 135 . Information from the sensors 130 and 135 can be merged with one another in a step 315 . In this way, weaknesses in one of the sensors 130, 135 can be compensated for by strengths in the respective other sensor 130, 135. An incorrect measurement can be suppressed in an improved manner. Different types of information can complement each other.

In a step 320, the first object 220 can be determined, which can be hidden at least partially and/or at least for the second sensor 135 behind the second object 230. The object 220 can be determined more precisely, for example in terms of size, speed, direction of travel or material. In a step 325, the probable trajectory 225 of the first object 220 can be determined. The anticipated trajectory 210 of the vehicle 105 can also be determined.

In a step 330 it can be determined on the basis of the probable trajectories 210, 225 whether a collision of the vehicle 105 with the first object 220 is imminent. This can be the case in particular when trajectories 210, 225 run in such a way that the distance between vehicle 105 and first object 220 falls below a predetermined minimum. A location where this occurs can be designated as a collision location 235 .

If a collision is to be feared, then in a step 335 a signal can be output on board vehicle 105 that indicates the impending collision. The quality or intensity of the signal can reflect the impending danger. For example, the closer the vehicle 105 is to the collision location 235 or the less time is left for the vehicle 105 to reach the collision location 235 , the more urgent the signal may be.

In a step 340 an automatic intervention in a control of the vehicle 105 can also take place. The intervention can relate to a longitudinal and/or lateral control of the vehicle 105 . For example, the vehicle 105 can be controlled to brake, accelerate, or evade. In particular, an attempt can be made to avoid collision location 235.

Reference List

100

system

105

vehicle

110

contraption

115

environment

120

driver

125

processing facility

130

first sensor, radar sensor

135

second sensor, infrared sensor

140

dispenser

145

acoustic output device

150

display device

205

first driveway

210

projected trajectory of the vehicle

215

second driveway

220

first object, further vehicle

225

expected trajectory of the first object or further vehicle

230

second object, tree

235

collision location

240

radar lobe

300

procedure

305

Scan the surroundings using a radar sensor

310

Scan the environment using an infrared sensor

315

merge information

320

determine object

325

determine trajectories

330

upcoming collision?

335

output warning

340

Intervention in vehicle control

Claims (14)

Method (300) for collision avoidance of a vehicle (105), comprising the following steps: (305, 310) scanning a lateral environment (115) of the vehicle (105); (320) detection of a first object (220) moving relative to the surroundings (115), which is at least partially covered by a second object (230); (325) determining an expected trajectory (225) of the first object (220); determining (325) an expected trajectory (210) of the vehicle (105); and determining (330), based on the trajectories (210, 225), whether a collision of the vehicle (105) with the first object (220) is imminent. Method (300) according to claim 1 , wherein the first object (220) has a predetermined minimum stand from the expected trajectory (210) of the vehicle (105). Method (300) according to claim 1 or 2 , wherein a portion of the first object (220) is obscured by the second object (230). A method (300) according to any one of the preceding claims, wherein the first object (220) is detected for a predetermined duration and is occluded by the second object (230) for a portion of the duration. Method (300) according to one of the preceding claims, wherein on the basis of the trajectories (210, 225) a probability is determined with which a collision of the vehicle (105) with the first object (220) is imminent. Method (300) according to one of the preceding claims, wherein a signal is provided (335) which indicates a time interval up to an imminent collision. A method (300) according to any one of the preceding claims, wherein the vehicle (105) is controlled (340) to avoid the collision. Device (110) for collision avoidance of a vehicle (105), the device (110) comprising the following elements: a first sensor (130) for scanning a lateral environment (115) of the vehicle (105); an output device (140) for providing a signal; and a processing device (125) which is set up to detect a first object (220) which is moving relative to the surroundings (115) and which is at least partially covered by a second object (230); determine an expected trajectory (225) of the first object (220); determine an expected trajectory (210) of the vehicle (105); on the basis of the trajectories (210, 225) to determine whether a collision of the vehicle (105) with the first object (220) is imminent; and in this case to provide a signal. Device (110) according to claim 8 , wherein the first sensor (130) emits waves and receives reflected waves, wherein the second object (230) is at least partially transparent to the emitted waves. Device (110) according to claim 9 , wherein the first sensor (130) is adapted to emit waves of different frequencies in order to at least partially penetrate different materials. Device (110) according to any one of claims 9 or 10 , wherein the first sensor (130) comprises a radar sensor. Device (110) according to any one of Claims 8 until 11 , further comprising a second sensor (135) for scanning the lateral environment (115), wherein the second sensor (135) comprises an infrared sensor; wherein the first object (220) is detected based on readings of both sensors (130, 135). Device (110) according to any one of Claims 8 until 12 , further comprising a display device (150) for providing a graphical representation of the first object (220). Vehicle (105) comprising a device (110) according to one of Claims 8 until 13 .

Images