Patent Application
20230249675
The present invention relates to a method for ascertaining a replacement trajectory, a computer program product, a parking assistance system and a vehicle having a parking assistance system.
Parking assistance systems are known that can autonomously retrace a trained trajectory with a vehicle. In these, the trajectory is first trained, that is, a user of the vehicle drives the trajectory to be trained manually, wherein the parking assistance system or another system records the trajectory. At a later time, the user can then have the trained trajectory retraced by the parking assistance system.
This can be problematic if environmental conditions have changed in the meantime, in particular if obstacles are found in the area of the trained trajectory. This often leads to a termination of the retracing maneuver.
Document DE 10 2017 115 988 A1 describes a method for the automated operation of a vehicle, in which a trajectory is provided and an object is detected in a region corresponding to the trajectory. The object detected in the region is classified and the motion path of the trajectory is modified depending on the classification of the detected object.
Against this background, one object of the present invention is to improve the operation of a vehicle.
According to a first aspect, a method for ascertaining a replacement trajectory for a vehicle is proposed, which is operable by means of a parking assistance system in an autonomous driving mode. In a first step, a predefined trajectory is received that comprises at least a first section and a second section that are linked to one another at a travel direction turning point, wherein a travel direction of the first section is different from a travel direction of the second section. In a second step, a sensor signal indicative of the surroundings of the vehicle is received. In a third step, an obstacle in the environment is detected based on the received sensor signal. In a fourth step, at least one collision point is calculated which is a point on the predefined trajectory, at which a collision occurs between the vehicle and the obstacle. The calculation is based on the predefined trajectory, the detected obstacle and a vehicle geometry of the vehicle. In a fifth step, the replacement trajectory is ascertained based on at least one calculated collision point. The replacement trajectory connects a starting point on the predefined trajectory located before the collision point to an end point on the predefined trajectory located after the collision point, avoiding the collision.
This method has the advantage that an obstacle which prevents safe continuation along the predefined trajectory, in particular when this is in the region of the travel direction turning point, can be bypassed according to the ascertained replacement trajectory. This means that a corresponding retracing maneuver can be carried out successfully and does not have to be aborted. The proposed method is carried out in particular by the parking assistance system of the vehicle.
The parking assistance system, which can also be referred to as a driver assistance system, is configured in particular for partially autonomous or fully autonomous driving of the vehicle. Partially autonomous driving is understood as meaning, for example, that the parking assistance system controls a steering apparatus and/or an automatic speed level system. Fully autonomous driving is understood as meaning, for example, that the parking assistance system also additionally controls a drive device and a braking device. The parking assistance system may be implemented in the form of hardware and/or in the form of software. In the case of an implementation in the form of hardware, the parking assistance system may be, for example, in the form of a computer or a microprocessor. In the case of an implementation in the form of software, the parking assistance system may be in the form of a computer program product, a function, a routine, part of a program code or an executable object. In particular, the parking assistance system may be in the form of part of a superordinate control computer of the vehicle, for example an ECU (Engine Control Unit).
The vehicle is, for example, an automobile or even a truck. Preferably, the vehicle comprises a number of sensor units which are configured to capture the driving state of the vehicle and to capture an environment of the vehicle. Examples of such sensor units of the vehicle are image acquisition devices, such as a camera, a radar (radio detection and ranging) or a lidar (light detection and ranging), ultrasonic sensors, location sensors, wheel angle sensors and/or wheel speed sensors. The sensor units are each configured to output a sensor signal, for example to the parking assistance system which carries out the partially autonomous or fully autonomous driving on the basis of the captured sensor signals.
The predefined trajectory is preferably a trained trajectory. For example, the parking assistance system or another system of the vehicle is configured to record and store a manually driven trajectory in a training mode. This involves, for example, recording various sensor signals, which characterize a driving condition of the vehicle, such as a speed, a position, a steering angle and the like, as uniquely as possible. In addition, sensor signals are recorded by environment sensors of the vehicle, which enable, for example, an image of the environment of the vehicle, in particular a position of obstacles in the environment. By playing back the driving state of the vehicle time-synchronously, i.e. by repeating it, the trained trajectory can be retraced.
For example, it is provided that a user starts a retracing maneuver by means of an input device, wherein the user selects the trajectory to be traced from a number of predefined trajectories or the parking assistance system proposes a suitable trajectory to the user based on a current position and orientation of the vehicle.
In this case, the predefined trajectory comprises at least one travel direction turning point, at which the travel direction of the vehicle changes. Accordingly, the travel direction in a first section before the travel direction turning point is opposite to a travel direction in a second section after the travel direction turning point. The travel direction can be determined, for example, by means of a direction of rotation of a wheel of the vehicle, wherein the direction of rotation of the wheel differs on the two sections, i.e. is first counter-clockwise and then clockwise.
To retrace the predefined trajectory, it is desirable to take current environment sensor data into account. Therefore, the parking assistance system receives a sensor signal that is indicative of the surroundings. The parking assistance system can, for example, receive this directly from one or more environment sensors of the vehicle and combine multiple sensor signals of different environment sensors, or else the parking assistance system receives the sensor signal already in a preprocessed state, for example in the form of a digital environment map, in which detected obstacles in the environment are indicated.
Then, an obstacle in the environment is detected based on the received sensor signal. The detection of the obstacle comprises, for example, detecting coordinates of the obstacle or an outline of the obstacle in a coordinate system of the vehicle, detecting a geometry of the obstacle, detecting a type of the obstacle and the like. Put another way, the obstacle is classified. It is also possible to ascertain whether it is a static obstacle or a moving obstacle. A static obstacle does not change its position during the retracing maneuver or only within the range of a measurement inaccuracy, and a mobile obstacle moves or could move.
Then at least one collision point is calculated, which is a point on the predefined trajectory, at which a collision occurs between the vehicle and the obstacle. The calculation point is calculated based on the predefined trajectory, the detected obstacle and a vehicle geometry of the vehicle. The vehicle geometry of the vehicle is predefined in particular and comprises, for example, a geometric model with a plurality of edges and surfaces. For example, a collision point is a point at which, if the vehicle were to travel along the specified trajectory, the vehicle would touch the obstacle. This retracing can be simulated by the parking assistance system, for example. Each point on the predefined trajectory, at which the vehicle is in contact with or overlapping the obstacle, is calculated as a collision point, for example. A collision point can also exist if a predefined minimum distance, for example a safety distance, to the obstacle is undershot.
After the at least one collision point has been calculated, the replacement trajectory is ascertained based on the at least one calculated collision point. The replacement trajectory connects a starting point on the predefined trajectory, which is located before the collision point, to an end point on the predefined trajectory, which is located after the collision point, avoiding the collision. The replacement trajectory thus replaces a section in the predefined trajectory such that the target position of the predefined trajectory is attainable without a collision or a termination of the retracing maneuver occurring due to the obstacle. Preferably, the length of the replacement trajectory is minimized, so the replacement trajectory is as short as possible. This keeps a deviation from the predefined trajectory to a minimum.
According to one embodiment, the parking assistance system causes the vehicle to drive along the ascertained replacement trajectory. In particular, it is understood that the parking assistance system as described above operates the vehicle in a semi-autonomous or fully autonomous driving mode by generating and outputting the corresponding control commands and the like.
According to a further embodiment, the starting point of the replacement trajectory is located on the first section and the end point of the replacement trajectory is located on the second section.
In this embodiment, the replacement trajectory replaces a section of the predefined trajectory comprising the travel direction turning point. This is the case in particular if the obstacle is arranged in a region of the travel direction turning point, so that for example the travel direction turning point can no longer be reached without collisions.
According to a further embodiment, a distance of the obstacle from the travel direction turning point is less than a predefined limit value.
The predefined limit value depends, for example, on the dimensions of the vehicle and a maximum attainable steering angle. For example, for a small car, the limit value may be 3 m, 4 m, or 5 m. For a truck, the limit may be 7 m, 8 m or 9 m, for example.
In embodiments, the limit value for the distance of the obstacle from the travel direction turning point depends on a relative position of the obstacle to the first section and the second section. For example, the obstacle can lie on the specified trajectory, in which case the limit is set relatively large. In another example, the obstacle may not be located on the predefined trajectory behind the travel direction turning point. The limit value can then be set relatively small.
According to a further embodiment, the replacement trajectory comprises at least one further travel direction turning point.
It can also be said that the replacement trajectory replaces a travel direction change section of the predefined trajectory.
According to a further embodiment the replacement trajectory has a section which lies in an overlapping section on the predefined trajectory and the travel direction of which is opposite to the travel direction of the predefined trajectory in the overlapping section.
It may be the case that the obstacle is detected during the retracing of the predefined trajectory at a time when it is not possible to divert from the given trajectory to the left or right, for example because the obstacle was detected too late. Then, for example, the replacement trajectory can initially comprise backtracking on the predetermined trajectory to bring the vehicle to an improved starting position from which the obstacle can be bypassed without collisions.
According to a further embodiment, a maximum offset of the replacement trajectory with respect to the predefined trajectory is less than a predefined maximum offset.
For example, the offset is defined as the shortest distance of a point on the replacement trajectory to the predefined trajectory. The offset can also be determined taking into account the vehicle geometry.
According to a further embodiment, the replacement trajectory has at least three travel direction turning points.
For example, in particularly complicated cases, in which a deviation from the predetermined trajectory is barely possible, a maneuvering of the vehicle can be helpful to avoid the obstacle by means of the replacement trajectory.
According to a further embodiment, a distance of the starting point of the replacement trajectory to the collision point is less than a predefined maximum distance.
This ensures that the deviation from the specified trajectory is not made too early. The predefined maximum distance can depend in particular on the vehicle geometry and a maximum steering angle. For example, the predefined maximum distance is 3 m, 4 m, 5 m, 6 m, 7 m, 8 m, 9 m, or even 10 m.
According to a further embodiment of the method, the length of the replacement trajectory is less than a predefined maximum length.
According to a second aspect, what is proposed is a computer program product that comprises commands that, when the program is executed by a computer, prompt said computer to perform the method according to the first aspect.
A computer program product, such as a computer program means, can be provided or supplied as a storage medium, such as a memory card, USB stick, CD-ROM, DVD, or else in the form of a downloadable file from a server to a network. This may take place, for example, in a wireless communication network by transmitting a corresponding file containing the computer program product or the computer program means.
According to a third aspect, a parking assistance system for a vehicle is proposed. The parking assistance system, which can also be referred to as a driver assistance system, is configured for automatically driving the vehicle along a trajectory. The parking assistance system comprises a receiving unit for receiving a predefined trajectory and for receiving a sensor signal indicative of surroundings of the vehicle. The predefined trajectory comprises at least a first section and a second section that are linked to one another at a travel direction turning point, wherein a travel direction of the first section is different from a travel direction of the second section. The parking assistance system further comprises a detection unit for detecting an obstacle in the surroundings depending on the received sensor signal, and a calculation unit for calculating at least one collision point, which is a point on the predefined trajectory at which a collision occurs between the vehicle and the obstacle, depending on the predefined trajectory, the detected obstacle and a vehicle geometry of the vehicle. In addition, the parking assistance system has an ascertaining unit for ascertaining a replacement trajectory based on the at least one calculated collision point, wherein the replacement trajectory connects a starting point on the predefined trajectory, which is located before the collision point, to an end point on the predefined trajectory, which is located after the collision point, while avoiding the collision.
The parking assistance system is preferably operated with the method of the first aspect. The parking assistance system has the same advantages that have been described for the method of the first aspect. The embodiments and features described for the proposed method apply accordingly to the proposed parking assistance system.
The parking assistance system is designed in particular for partially autonomous or fully autonomous driving of the vehicle. Partially autonomous driving is understood as meaning, for example, that the parking assistance system controls a steering apparatus and/or an automatic speed level system. Fully autonomous driving is understood as meaning, for example, that the parking assistance system also additionally controls a drive device and a braking device. The parking assistance system and/or any unit of the parking assistance system, for example the reception unit, the detection unit, the calculation unit and/or the ascertainment unit, may be implemented in the form of hardware and/or in the form of software. In the case of an implementation in the form of hardware, the parking assistance system or the respective unit may be designed for example as a computer or as a microprocessor. In the case of an implementation in the form of software, the parking assistance system or the respective unit may be designed as a computer program product, as a function, as a routine, as part of a program code or as an executable object. In particular, the parking assistance system or the respective unit may be in the form of part of a superordinate control computer or control system of the vehicle, for example an ECU (Engine Control Unit).
According to one embodiment of the parking assistance system, this is configured for automatic driving of the vehicle along the replacement trajectory.
A further aspect proposes a vehicle with a parking assistance system according to the third aspect.
The vehicle is, for example, an automobile or even a truck. Preferably, the vehicle comprises a number of sensor units which are configured to capture the driving state of the vehicle and to capture an environment of the vehicle. Examples of such sensor units of the vehicle are image acquisition devices, such as a camera, a radar (radio detection and ranging) or a lidar (light detection and ranging), ultrasonic sensors, location sensors, wheel angle sensors and/or wheel speed sensors. The sensor units are each configured to output a sensor signal, for example to the parking assistance system which carries out the partially autonomous or fully autonomous driving on the basis of the captured sensor signals.
Further possible implementations of the invention also comprise not explicitly mentioned combinations of features or embodiments described above or below with regard to the exemplary embodiments. A person skilled in the art will in this case also add individual aspects as improvements or additions to the respective basic form of the invention.
Further advantageous configurations and aspects of the invention are the subject of the dependent claims and of the exemplary embodiments of the invention that are described below. The invention is explained in more detail below on the basis of preferred embodiments with reference to the accompanying figures.
Identical or functionally identical elements have been provided with the same reference signs in the figures, unless stated otherwise.
The parking assistance system 110 is designed to ascertain a replacement trajectory ET (see
The predefined trajectory VT is, for example, a trained trajectory and leads from a start position SP to an end position EP. In this example, the predefined trajectory VT comprises two travel direction turning points WP, at each of which the travel direction DIR of the vehicle 100 changes, as is evident from the arrows DIR.
In this first scenario, during a retracing procedure an obstacle 210 is arranged on the predefined trajectory VT in a first section A1 before the first travel direction turning point WP of the predefined trajectory VT. A collision point KP is calculated, at which the vehicle 100 collides with the obstacle 210 if it continues to drive on the predefined trajectory VT.
Therefore, a replacement trajectory ET is ascertained. This starts at a starting point ET1 on the predefined trajectory VT just before the obstacle 210. The starting point ET1 is also a travel direction turning point WP, that is, the travel direction DIR of the vehicle 100 is reversed at this point. The replacement trajectory ET follows a rolling path in a single section to an end point ET2 on the predefined trajectory VT, which lies on a second section A2 of the predefined trajectory VT. The travel direction DIR of the vehicle 100 at the end point ET2 corresponds to the travel direction DIR of the second section A2 of the predefined trajectory VT.
The remaining part of the predefined trajectory VT is free of obstacles 210 and can therefore be retraced as intended.
The special feature in this example is that the replacement trajectory ET still departs from the predefined trajectory VT in the first section A1 before the travel direction turning point WP. The replacement trajectory ET in this example is similar to that described in
The obstacle 210 has a distance Dmin from the predefined trajectory VT. The distance Dmin is below a minimum distance that must be maintained from an obstacle. Due to the vehicle geometry, for example, a collision would result at the collision point KP despite the distance Dmin, which is why the replacement trajectory ET is required. In this example, the collision point KP is only calculated when the vehicle 100 is already quite close to the obstacle 210, which is why the vehicle 100 must initially backtrack slightly. Therefore, the starting point ET1 corresponds to a travel direction turning point WP. An overlapping section A3 is formed, in which the predefined trajectory VT and the replacement trajectory ET overlap each other, but each having an opposite travel direction DIR. Due to the limitation LIM of the drivable region, a replacement trajectory ET as shown in the example of
In this example, the ascertaining unit 118 outputs the replacement trajectory ET to a unit (not shown) outside the parking assistance system 110. Alternatively, it can be provided that the parking assistance system 110 causes the vehicle 100 to drive along the ascertained replacement trajectory ET.
The parking assistance system 110 and/or any unit of the parking assistance system 110, for example the reception unit 112, the detection unit 114, the calculation unit 116 and/or the ascertainment unit 118, may be implemented in the form of hardware and/or in the form of software. In the case of an implementation in the form of hardware, the parking assistance system 110 or the respective unit 112, 114, 116, 118 may be designed for example as a computer or as a microprocessor. In the case of an implementation in the form of software, the parking assistance system 110 or the respective unit 112, 114, 116, 118 may be designed as a computer program product, as a function, as a routine, as part of a program code or as an executable object. In particular, the parking assistance system 110 or the respective unit 112, 114, 116, 118 may be configured as part of a higher-level control computer or control system (not shown) of the vehicle 100.
This method is advantageously carried out with the parking assistance system 110 of
Although the present invention has been described on the basis of exemplary embodiments, it may be modified in many ways.
| Number | Date | Country | Kind |
|---|---|---|---|
| 10 2020 117 773.9 | Jul 2020 | DE | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP2021/067802 | 6/29/2021 | WO |
