Method for evaluating reception information of a lidar sensor, and lidar sensor
The method enhances LiDAR sensor capabilities by detecting objects in close proximity through echo analysis, overcoming the limitations of long-range sensors in precise near-range measurements.
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- VALEO SCHALTER & SENSOREN GMBH
- Filing Date
- 2026-01-21
- Publication Date
- 2026-07-30
AI Technical Summary
Long-range LiDAR sensors are incapable of performing precise measurements at close range, particularly in determining the presence of objects in the immediate vicinity.
A method for evaluating received LiDAR information by comparing echo information with a reference echo, allowing detection of objects in close proximity without precise distance measurement, utilizing the known characteristics of the reference echo to analyze the presence of objects based on temporal and parameter analysis of echo information.
Enables accurate detection of objects in the immediate vicinity of the LiDAR sensor, utilizing raw echo information for initial presence assessment, which can then be further processed for precise distance determination.
Smart Images
Figure EP2026051366_30072026_PF_FP_ABST
Abstract
Description
[0001] 2024PF01991
[0002] 1
[0003] Method for evaluating received information from a LIDAR sensor, as well as LIDAR sensor
[0004] One aspect of the invention relates to a method for evaluating received information from a LiDAR sensor. Another aspect of the invention relates to a LiDAR sensor for a vehicle.
[0005] LiDAR sensors are well known for mapping the area around a vehicle. However, LiDAR sensors designed for long-range measurements are not, or only partially, capable of performing precise measurements at close range, particularly measuring precise distances to objects in the environment. A LiDAR sensor typically has a housing with a window through which laser pulses are emitted by a transmitter unit of the LiDAR sensor and transmitted into the surrounding area. Reflected pulses also pass through this window and are received by a receiver unit of the LiDAR sensor. The window is positioned at a fixed distance from the transmitter unit. It is also known that a single emitted pulse can have multiple echoes, i.e., multiple reflected signals.This occurs due to different objects being at varying distances, each of which receives a portion of the transmitted pulse and is reflected.
[0006] The object of the present invention is to provide a method and a LIDAR sensor with which, or in which, received information, in particular with regard to detections in the immediate vicinity of the transmitting unit, can be evaluated in an improved manner.
[0007] This task is solved by a method and a LIDAR sensor according to the independent claims.
[0008] One aspect of the invention relates to a method for evaluating received information from a LiDAR sensor. The received information acquired by the LiDAR sensor is thus evaluated. In particular, the method comprises the following steps: 2024PF01991
[0009] 2
[0010] In particular, sending a transmission pulse with a transmitter unit of the LIDAR sensor;
[0011] In particular, receiving echo information as reception information based on the transmit pulse with a receiving unit of the LIDAR sensor, wherein the echo information characterizes a reflection of the transmit pulse at at least one object, comparing the echo information with a reference echo that characterizes a reflection, in particular at least a part, of the transmit pulse at a window of the LIDAR sensor through which the transmit pulse is emitted into the environment of the LIDAR sensor,
[0012] Evaluating the echo information as received information such that an object in the environment is detected in a close proximity to the LIDAR sensor if the echo information deviates from the reference echo and the echo information is received at least partially within a time window after the transmission of the transmit pulse in which the reference echo is expected.
[0013] This method now makes it possible to more accurately evaluate and analyze received information from the LiDAR sensor, even at close range. A particular advantage of this method is the ability to analyze complex received information and detect the presence of an object in the immediate vicinity. The method's primary benefit is its ability to easily and precisely determine whether an object is present in this immediate area. Therefore, determining the exact distance of this potential object from the LiDAR sensor is initially irrelevant. The crucial factor is the detection of the object's presence. The precise distance of this object from the LiDAR sensor is of secondary importance for the proposed method.In this context, it's broadly about the general assessment of whether an object is present in the immediate vicinity. This can be done very accurately even without precise distance measurement, based on the proposed procedure steps. The method cleverly utilizes the fact that a reference echo, which is a reflection of part of the transmitted pulse at the LiDAR sensor's window, is very precisely known, and a corresponding expectation can be assumed. With this virtually constant and precisely known information, which can then always be expected as such, received information can be analyzed very accurately, in addition to this reference 2024PF01991.
[0014] 3
[0015] The echo also contains at least one other echo. This makes it easy to recognize that an object is present in the immediate vicinity, especially in such situations. Furthermore, selecting this echo information is particularly advantageous because the content of the reference echo is known, allowing the rest of the echo information to be recognized and analyzed.
[0016] Especially when the echo information lies at least partially within a time window after the transmission of the transmit pulse, in which the reference echo is also expected, this analysis can be performed and it may be concluded that, in addition to the reference echo, at least one further echo is present in the echo information.
[0017] The proposed method makes it possible, particularly with long-range LiDAR sensors, to at least determine whether an object is present in the immediate vicinity based on the proposed analysis. In this initial assessment, precise distance determination plays no role, or at least a subordinate one.
[0018] Especially with a LIDAR sensor designed as a long-range LIDA sensor, a corresponding rough analysis can be carried out in the near range to determine whether an object is present or not.
[0019] The entire process can be performed by the LiDAR sensor itself. It is also possible for some steps, such as comparing and / or evaluating, to be performed externally. In this case, a system is formed that includes the LiDAR sensor and at least one external unit.
[0020] In one embodiment, the temporal length of the echo information is compared with a known temporal reference length of the reference echo. If the temporal length of the echo information is longer than the temporal reference length, the echo information is recognized as multi-echo information consisting of several, in particular temporally overlapping, echoes. This multi-echo information then includes the reference echo and at least one further echo as a reflection from the object in the immediate vicinity. In particular, an analysis of the respective temporal length and the temporal position along a timeline is thus particularly accurate in order to recognize whether such a 2024PF01991
[0021] 4
[0022] Another echo is present in the echo information, thus indicating that an object is present in the immediate vicinity.
[0023] The echo information is received as a continuous signal in terms of its temporal length. Therefore, the echo information is not such that two different echoes, for example, the reference echo and another echo from a nearby object, are temporally separated and received as separate pulses and thus as individual echoes. Instead, viewed in the timeline, the reference echo and at least one other echo originating from a nearby object overlap, at least partially. In particular, there is no complete overlap, so there is no perfectly superimposed alignment of the reference echo and the other echo.
[0024] In one embodiment, the duration of the echo information is characterized based on a known or expected start time of the echo information. The start time of the echo information is characterized, in particular, by the start time of the reference echo. The duration of the echo information is further characterized by an end time of the echo information. This end time can be generated, for example, by an evaluation unit of the LiDAR sensor. This is particularly relevant if the evaluation unit is capable of detecting or analyzing an echo end and providing this information. For example, this could be the case with a suitable imager of the LiDAR sensor that can perform such an evaluation and provide such an echo end as information, and which serves as an example of an evaluation unit.
[0025] The duration of the echo information is therefore determined in particular as the time between the start time and the end time.
[0026] In one embodiment, the temporal reference length is characterized by initial measurements in the echo information. The echo information can contain second measurements that differ in value from the initial measurements. These second measurements characterize at least one further echo. In particular, the temporal length of the further echo is characterized, and especially determined, by these second measurements. Such a measurement analysis of the entire echo information allows for a relatively precise analysis of the proportion formed by the reference echo and the proportion formed by the further echo. Thus, it can be provided that 2024PF01991
[0027] 5
[0028] Lower measured values characterize the reference echo, and higher measured values characterize the further echo, which originates as a reflection from an object in the immediate vicinity.
[0029] In one embodiment, the echo information is analyzed with respect to at least one characterizing parameter and compared with a reference parameter of the reference echo. If the parameter values differ, it is recognized that an object is present in the environment within close proximity to the transmitting unit and thus also to the LIDAR sensor.
[0030] Such a parameter analysis can be performed in addition to or instead of the scenario mentioned above, in which an evaluation unit can, for example, generate and provide an echo end as information.
[0031] At least one of the following parameters can be evaluated as a characterizing parameter: the pulse width of the echo information, the pulse area of the echo information, or the ratio of the pulse amplitude of the echo information to its duration. The pulse width, in this context, could, for example, be the full width at half maximum (FWHM). These specific characterizing parameters enable a very precise analysis of the echo information and thus also the selection of a reference echo and at least one other echo that may be present.
[0032] In one embodiment, the near range is defined as an area of the surroundings less than or equal to two meters, and in particular less than or equal to one meter, from the LIDAR sensor. This exemplary range is particularly critical for long-range LIDAR sensors with regard to the accurate detection, especially the determination of distances to objects.
[0033] In one embodiment, the evaluation of the echo information is performed before determining the distance of any object that may be present in the vicinity of the LIDAR sensor. Thus, the entire analysis procedure aims to first establish, in general terms, whether an object is present in the vicinity or not, regardless of the specific and precise distance that potential object may have from the LIDAR sensor. 2024PF01991
[0034] 6
[0035] In one embodiment, it is provided that the echo information, which is taken into account for the assessment of an object in the immediate vicinity, is provided as coarse information that is insufficient for determining a distance.
[0036] In this context, raw data representing the echo information is provided. This is particularly advantageous for long-range LiDAR sensors, as they can only provide very limited or no accurate distance measurement of such close objects based on this raw data.
[0037] The proposed method cleverly allows for the use of this raw, unprocessed information, particularly for an initial assessment of whether an object is present in the immediate vicinity. A key advantage of this method is its ability to utilize the raw, echo-based information—which is generally unusable at this stage with long-range LiDAR sensors—to generate general information about the presence of an object in the immediate vicinity.
[0038] In one embodiment, the echo information is further processed after evaluation—that is, after analyzing whether an object is present in the immediate vicinity—and the processed echo information is then used to determine a distance. This processing thus enables the echo information, initially available only as raw data, to be used as a basis for a potentially more precise distance determination. This distance determination can then also be used for objects that are further away and / or for the at least one potentially present object in the immediate vicinity, allowing for a more accurate or sufficiently accurate distance value to be determined.
[0039] Another aspect of the invention relates to a LiDAR sensor for a vehicle. The LiDAR sensor can also be referred to as a vehicle LiDAR sensor. The LiDAR sensor is, in particular, a long-range LiDAR sensor. This means, in particular, that it is designed to detect objects at a distance greater than 50 meters, and especially greater than 100 meters, from the LiDAR sensor. The detection range can extend to a maximum of several hundred meters. In particular, the LiDAR sensor is configured to perform the method according to the aspect mentioned above.
[0040] 7
[0041] or an advantageous embodiment thereof. In particular, the method is carried out using the LIDARF sensor.
[0042] The LiDAR sensor preferably comprises a transmitter and a receiver. The transmitter can be a laser module. A transmit pulse can be emitted as a laser pulse. Preferably, the wavelength of such a transmit pulse is in the infrared range. The wavelength can preferably be between 800 nanometers and 1600 nanometers, in particular, for example, 850 nanometers, 905 nanometers, or 1550 nanometers.
[0043] The LiDAR sensor can also include an optical system. This system can include at least one mirror with which the transmitted pulse is directed through the window mentioned above from the housing into the surroundings. This mirror, which can also be a rotating mirror, can also direct received information, in particular an echo, which represents at least a portion of the transmitted pulse reflected in the surroundings, to the receiving unit. The LiDAR sensor specifically includes a housing. This housing has a window, as described above. This window is designed to transmit the transmitted pulse into the surroundings and to allow received information to pass through into the housing. The transmitting unit can also include surface emitters, in particular VCSELs (vertical cavity surface emitting lasers).
[0044] Another aspect of the invention relates to a vehicle with at least one LIDAR sensor according to the above-mentioned aspect or an advantageous embodiment thereof.
[0045] Another aspect of the invention relates to a method for detecting the environment of a vehicle with a LIDAR sensor, in particular the vehicle itself.
[0046] The method preferably comprises the process steps as described above for the method of evaluating received information from a LiDAR sensor. Based on this evaluated received information, it is determined whether an object is present in the vicinity of the LiDAR sensor or not, in particular without yet performing a specific distance determination of such a potential object in the vicinity of the LiDAR sensor. 2024PF01991
[0047] 8
[0048] Advantageous embodiments of the above-mentioned method can also be considered as advantageous embodiments for the method of detecting the environment of the LIDAR sensor.
[0049] Exemplary embodiments of the invention are explained in more detail below with reference to schematic drawings. These show:
[0050] Fig. 1 shows a schematic representation of an embodiment of the vehicle according to the invention with an embodiment of a LIDAR sensor according to the invention; and
[0051] Fig. 2 is a schematic diagram showing an example of schematically displayed echo information.
[0052] In the figures, identical or functionally equivalent elements are given the same reference symbols.
[0053] Figure 1 shows a schematic representation of an embodiment of a vehicle 1. The vehicle 1 can be a passenger car, a truck, or a bus. The vehicle 1 has an embodiment of a LiDAR sensor 2. The LiDAR sensor 2 can also be referred to as a vehicle-mounted LiDAR sensor. The LiDAR sensor 2 is permanently installed on the vehicle 1. It is designed to detect the area 3 surrounding the vehicle 1. The LiDAR sensor 2 is, in particular, a long-range LiDAR sensor.
[0054] The LiDAR sensor 2 comprises a transmitter 4 and a receiver 5. The LiDAR sensor 2 also includes, in particular, an optical system 6. This system may include a rotating mirror. A transmit pulse emitted as a laser pulse from the transmitter 4 is deflected by the optical system 6 to a window 7 of the LiDAR system and emitted through the window 7 into the surroundings 3. Received information from the surroundings enters the interior of a housing 8 of the LiDAR sensor 2 through the window 7 and is transmitted, in particular, by the optical system 6 to the receiver 5.
[0055] In one embodiment, the LIDAR sensor 2 preferably also includes an evaluation unit 9. The evaluation unit 9 can be a processing unit or have a processing unit. The processing unit can be a microprocessor. 2024PF01991
[0056] 9
[0057] Evaluation unit 9 can, for example, be part of the receiver unit 5 or be separate from it. Evaluation unit 9 can also be part of a control unit for the LIDAR sensor 2. The control unit manages the generation and transmission of the transmit pulses from the transmitter unit 4. In particular, the optical system 6, especially a rotating mirror, can also be controlled by the control unit.
[0058] Window 7 is transparent to at least the wavelength of the transmitted pulse.
[0059] The LIDAR sensor 2 makes it possible, in particular, to carry out a method for capturing the environment 3. Specifically, the LIDAR sensor 2 provides the possibility of carrying out a method for evaluating received information.
[0060] It is therefore possible to transmit a pulse P using the LIDAR sensor 2 and the transmitter 4. An echo E is then received as a receive signal. This echo E is received by the receiver 5 based on the pulse P. The echo E is at least partially a reflection of the pulse P from at least one object in the environment 3. The echo E is then compared with a reference echo RE, as shown schematically in the diagram in Fig. 2. The diagram is a time-amplitude diagram. The horizontal axis represents the time duration t and the vertical axis the amplitude A. The reference echo RE represents a reflection of a portion of the pulse P from the window 7.
[0061] For the reference echo RE, the echo length, and thus the duration of the corresponding received pulse, is known or can be expected from existing knowledge. Furthermore, the time and time interval over which this reference echo RE is received by receiver 5 after the transmission of the transmit pulse P is also known. This is because the distance between the transmit unit 4 and window 7 is permanently fixed and known. Based on the known propagation time of the transmit pulse P and this fixed distance of window 7 to both the transmit unit 4 and the receiver 5, the timing of this reference echo RE along the timeline is known and can therefore be predicted.
[0062] In the illustrated embodiment, an object 10 is also present in a near area 3a of the environment 3. This can be a static object.
[0063] 10
[0064] but it can also be a dynamic object. For example, object 10 could be another vehicle positioned close to vehicle 1. For example, object 10 could be moving close past vehicle 1.
[0065] If the transmitted pulse P, or in particular a further part thereof, also encounters this object 10 in the near field 3a, then a part of the transmitted pulse P is also reflected by the object 10 and received as part of the echo information. In this context, Fig. 2 shows an example of a further echo WE in the diagram, which characterizes the reflection of a part of the transmitted pulse P by the object 10.
[0066] In this embodiment, the entire echo information E is therefore formed from the reference echo RE and the further echo WE. Because the object 10 is located in the near field 3a and is detected, the two echoes overlap, namely the reference echo RE and the further echo WE.
[0067] Therefore, in such a configuration, object 10 can no longer be clearly differentiated in the echo information E. If, for example, the near range of the LIDAR sensor 2 is set such that window 7 is permanently visible in the first echo, which is characterized by the reference echo RE, this generates a simple pulse in the histogram shown in Fig. 2. If object 10 approaches further, or remains in this near range 3a, this actually generates a second pulse, namely the further echo WE shown schematically. However, this further echo WE is so close to the first pulse, i.e., the reference echo RE, that these two pulses, shown here as examples, are evaluated as the only echo information E and thus as the only pulse. This would subsequently lead to an incorrect distance determination and also an incorrect amplitude measurement.
[0068] Since, as described above, in such a constellation with echo information E the characteristic of the first echo and thus of the reference echo RE is known, further analysis and evaluation of the echo information E can take place, especially with regard to assessing whether an object 10 is present in the near field 3a or not.
[0069] This can occur, in particular, on the basis that the echo information E differs from the reference echo RE and that the echo information E is received, at least partially, within a time window after the transmission of the transmit pulse P, in which the reference alliance echo RE is expected. 2024PF01991
[0070] 11
[0071] It is possible that the temporal length of the echo information E is compared with a known temporal reference length of the reference echo RE, and if the temporal length of the echo information E is longer than the temporal reference length, the echo information E is recognized as multi-echo information consisting of several, in particular temporally overlapping, echoes. It is then also recognized that this echo information E, as multi-echo information, contains the reference echo RE and at least one further echo, here the further echo WE, as a reflection from object 10 in the near field 3a.
[0072] It is possible that the evaluation unit 9 is configured to provide information about an echo end of the echo information E. For this purpose, first and second measured values can be used, which characterize the reference echo RE on the one hand and the further echo WE on the other.
[0073] Particularly when an evaluation unit cannot evaluate and provide such information about an echo end of the echo information E, parameters characterizing the echoes can be evaluated additionally or instead. For example, the pulse width of the echo information E and / or the pulse area of the echo information E and / or a ratio of the pulse amplitude of the echo information E to the temporal duration of the echo information E can be evaluated.
[0074] It is particularly advantageous that the echo information E, which is considered for assessing an object 10 in the near field 3a, is provided as coarse information that is insufficient for distance measurement, and in particular that the echo information is provided as raw data. The evaluation of the echo information E with regard to the presence of an object 10 in the near field 3a is carried out, in particular, before determining a specific distance to an object in the environment 3, especially to an object 10 in the near field 3a. It is also possible that the echo information E is further processed and that the distance determination is then carried out based on the processed echo information E, and in particular, that this is only possible then. In particular, a distance of the object 10 to the vehicle 1, especially the LIDAR sensor 2, can then also be determined.
Claims
2024PF01991 12 Patent claims 1. Method for evaluating received information from a LIDAR sensor (2), comprising the following steps: Emitting a transmit pulse (P) with a transmitting unit (4) of the LIDAR sensor (2); Receiving an echo information (E) as reception information based on the transmit pulse (P) with a receiving unit (5) of the LIDAR sensor (2), wherein the echo information (E) characterizes a reflection of the transmit pulse (P) at at least one object, Comparing the echo information (E) with a reference echo (RE) that characterizes a reflection of the transmit pulse (P) at a window (7) of the LIDAR sensor (2), through which the transmit pulse (P) is emitted into the environment (3) of the LIDAR sensor (2), Evaluating the echo information (E) as received information such that an object (10) in the environment (3) is detected in a close range (3a) to the LIDAR sensor (2) if the echo information (E) differs from the reference echo (RE) and the echo information (E) is received at least partially within a time window after the transmission of the transmit pulse (P) in which the reference echo (RE) is expected.
2. Method according to claim 1, wherein the temporal length of the echo information (E) is compared with a known temporal reference length of the reference echo (RE), and if the temporal length of the echo information (E) is longer than the temporal reference length, the echo information (E) is recognized as multi-echo information consisting of several, in particular temporally overlapping, echoes, comprising the reference echo (RE) and at least one further echo (WE) as a reflection at the object (10) in the near field (3a).
3. The method of claim 2, wherein the echo information (E) is received as a continuous signal of the specified time length. 2024PF01991 13 4. Method according to claim 2 or 3, wherein a temporal length of the echo information (E) is determined on the basis of an initial time of the echo information (E), in particular a known or expected time, which is characterized by an initial time of the reference echo (RE), and an end time of the echo information (E) which is generated by an evaluation unit (9) of the LIDAR sensor (2).
5. Method according to claim 4, wherein the temporal reference length is characterized by first measured values in the echo information (E) and the echo information (E) has different second measured values, wherein the further echo (WE) is characterized by the second measured values, in particular the temporal length of the further echo (WE) is determined by these second measured values.
6. Method according to one of the preceding claims, wherein the echo information (E) is analyzed with respect to at least one characterizing parameter and is compared with a reference parameter of the reference echo (RE) and if the parameter values differ, it is recognized that the object (10) is present in the environment (3) in the near field (3a).
7. Method according to claim 6, wherein at least one of the characterizing parameters is evaluated from the pulse width of the echo information (E), the pulse area of the echo information (E), the ratio of pulse height of the echo information (E) to temporal length of the echo information (E).
8. Method according to one of the preceding claims, wherein the near area (3a) is considered to be an area of the environment (3) of less than or equal to 2 m, in particular less than or equal to 1 m, to the LIDAR sensor (2).
9. A method according to any of the preceding claims, wherein the evaluation of the echo information (E) is performed before determining the distance of any object (10) that may be present in the near field (3a) to the LIDAR sensor (2). 2024PF01991 14 10. Method according to one of the preceding claims, wherein the echo information (E) which is taken into account for the assessment of the object (10) in the near field (3a) is provided as coarse information, in particular raw data, which is insufficient for determining a distance between the object (10) and the LIDAR sensor (2).
11. Method according to claim 10, wherein the echo information (E) is processed and the processed echo information is used to determine the distance.
12. LIDAR sensor (2) comprising a transmitting unit (4) and a receiving unit (5), wherein the LIDAR sensor (2) is configured to perform a method according to one of the preceding claims.