Lidar chip and test system, point cloud generation method and readable storage medium

Abstract

This application discloses a lidar chip and testing system, a point cloud generation method, and a readable storage medium. The lidar chip includes a transmitting module, a receiving module, and a point cloud data processing module. The transmitting module emits a detection laser. The receiving module receives the echo laser formed by the reflection of the detection laser from a target object. The receiving module includes a receiving array and a time conversion unit. The receiving array includes multiple receiving units, and each receiving unit includes multiple receiving devices. The time conversion unit obtains the photon count value of each receiving device at different flight times based on the flight time of the photons received by each receiving device. The point cloud data processing module generates point cloud data based on the photon count values ​​of each receiving device at different flight times. One frame of point cloud data includes multiple point cloud pixels, each point cloud pixel corresponds to one receiving unit, and each point cloud pixel includes the ranging distance and reflectivity of the target object. This scheme can improve the dynamic range of a single point cloud pixel.

Description

Technical Field

[0001] This application relates to the field of lidar, and in particular to a lidar chip and testing system, a point cloud generation method, and a readable storage medium. Background Technology

[0002] Single-photon avalanche diode (SPAD) arrays represent the future direction of all-solid-state digital lidar. For a single SPAD, the photon detection efficiency (PDE) is typically 30%–40%, meaning that only 30%–40% of a photon hitting the SPAD is successfully recorded as a valid event, resulting in a low dynamic range for the obtained point cloud pixels. Summary of the Invention

[0003] This application provides a lidar chip and testing system, a point cloud generation method, and a readable storage medium, which can improve the dynamic range of a single point cloud pixel.

[0004] In a first aspect, embodiments of this application provide a lidar chip, comprising: a transmitting module for transmitting a detection laser; a receiving module for receiving an echo laser formed by the detection laser reflected from a target object; wherein the receiving module includes a receiving array and a time conversion unit, the receiving array including multiple receiving units, each receiving unit including multiple receiving devices, the time conversion unit being used to obtain the photon count value of each receiving device at different flight times based on the flight time of the photons received by each receiving device, the receiving devices being single-photon avalanche diodes; a testing module for testing the receiving array to determine the failed receiving devices in the receiving array; and a point cloud data processing module connected to the receiving module, used to replace the photon count value of the failed receiving device in each receiving unit with the expected value, and generate point cloud data based on the photon count value of each receiving device at different flight times, one frame of point cloud data including multiple point cloud pixels, each point cloud pixel corresponding to a receiving unit, and each point cloud pixel including the ranging distance and reflectivity of the target object.

[0005] In one or more embodiments, the point cloud data processing module includes: a merging unit, configured to replace the photon count value of the failed receiving device in each receiving unit with the expected value when the transmitting module emits a probe laser once, and to determine the photon count value of each receiving unit at the same flight time based on the photon count values ​​of all receiving devices in each receiving unit at the same flight time; a histogram accumulation unit, configured to accumulate multiple photon count values ​​of each receiving unit at different flight times to obtain an accumulated histogram, wherein each emission of the probe laser yields one photon count value for each receiving unit; and a point cloud generation unit, configured to determine the ranging distance and reflectivity corresponding to each point cloud pixel based on the accumulated histogram to generate the current frame point cloud data.

[0006] In one or more embodiments, the point cloud data processing module further includes a selection unit, and the merging unit includes multiple adders and multiple path selectors; each path selector corresponds one-to-one with a receiving device, the first input terminal of each path selector is used to obtain the photon count value of the corresponding receiving device, and the second input terminal of each path selector is used to obtain the expected value of the corresponding receiving device; the selection unit is connected to the control terminal of each path selector, and is used to select the second input terminal and output terminal of the path selector corresponding to the failed receiving device, or to select the first input terminal and output terminal of the path selector corresponding to the non-failed receiving device; the adders are used to accumulate the values ​​of the output terminals of the path selectors corresponding to the receiving devices in each receiving unit to obtain the photon count value of each receiving unit.

[0007] In one or more embodiments, the point cloud data processing module further includes an expected value configuration unit; the current frame point cloud is the Mth frame point cloud, the current frame point cloud includes a first point cloud pixel, and the first point cloud pixel corresponds to a first receiving unit; the expected value configuration unit is used to use an initial value as the expected value corresponding to the failed receiving device in the first receiving unit in the previous N frame point clouds, and to use the average of the photon count values ​​of the first receiving units in the M-1 frame point clouds as the expected value corresponding to the failed receiving device in the first receiving unit in the Mth frame point cloud; wherein, the transmitting module obtains a frame point cloud each time it transmits a probe laser, M and N are integers greater than 1, and M is greater than N.

[0008] Secondly, embodiments of this application provide a lidar chip testing system, including the lidar chip as described in the first aspect. The testing system further includes: a light source; the receiving module in the lidar chip further includes: a quenching circuit connected to the receiving array, used to quench and restore the receiving array; the testing module in the lidar chip includes: a first control unit connected to the light source, used to control the light source to emit light or stop emitting light; an observation and inspection unit connected to the first control unit and the quenching circuit, used to identify receiving devices that do not exhibit an avalanche effect as failed receiving devices in a first test mode, and used to identify receiving devices that exhibit an avalanche effect as failed receiving devices in a second test mode; wherein, the first control unit controls the light source to emit light in the first test mode, and the first control unit controls the light source to stop emitting light in the second test mode; a coordinate recording unit connected to the observation and inspection unit, used to record the position of the failed receiving device; a second control unit and a storage unit, the second control unit being connected to the coordinate recording unit and the storage unit being connected to the second control unit, the second control unit being used to write the position of the failed receiving device to the storage unit.

[0009] Thirdly, embodiments of this application provide a point cloud generation method applied to a lidar chip as described in the first aspect. The method includes: testing a receiving array to identify failed receiving devices in the array; obtaining photon count values ​​for each receiving device at different flight times; replacing the photon count values ​​of the failed receiving devices in each receiving unit with expected values, and determining the photon count value of each receiving unit at the same flight time based on the photon count values ​​of all receiving devices in each receiving unit at the same flight time; accumulating multiple photon count values ​​of each receiving unit at different flight times to obtain an accumulated histogram, wherein each emission of a detection laser yields one photon count value for each receiving unit; and determining the ranging distance and reflectivity corresponding to each point cloud pixel based on the accumulated histogram to generate point cloud data for the current frame.

[0010] In one or more embodiments, testing the receiving array to determine the failed receiving devices in the receiving array includes: placing the receiving array in a strong light environment and determining the receiving devices that do not exhibit an avalanche effect as failed receiving devices; placing the receiving array in a dark environment and determining the receiving devices that exhibit an avalanche effect as failed receiving devices.

[0011] In one or more embodiments, before replacing the photon count value of the failed receiving device in each receiving unit with the expected value, the method further includes: determining the location of the failed receiving device; and determining the receiving unit corresponding to the failed receiving device according to a lookup table of the location of the failed receiving device and the receiving device mapping relationship.

[0012] In one or more embodiments, the method further includes: the current frame point cloud is the Mth frame point cloud, the current frame point cloud includes first point cloud pixels, and the first point cloud pixels correspond to a first receiving unit; using an initial value as the expected value corresponding to the failed receiving device in the first receiving unit in the previous N frame point clouds, and using the average of the photon count values ​​of the first receiving units in the previous M-1 frame point clouds as the expected value corresponding to the failed receiving device in the first receiving unit in the Mth frame point cloud; wherein, each time a probe laser is emitted, a frame point cloud is obtained, M and N are integers greater than 1, and M is greater than N.

[0013] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed, implements the point cloud generation method as described in the third aspect.

[0014] The beneficial effects of this application are as follows: The lidar chip in this embodiment includes a transmitting module, a receiving module, and a point cloud data processing module. The transmitting module is used to transmit a detection laser. The receiving module is used to receive the echo laser formed by the detection laser reflected by the target object. The receiving module includes a receiving array and a time conversion unit. The receiving array includes multiple receiving units, and each receiving unit includes multiple receiving devices. The time conversion unit is used to obtain the photon count value of each receiving device at different flight times based on the flight time of the photons received by each receiving device. The point cloud data processing module is used to generate point cloud data based on the photon count values ​​of each receiving device at different flight times. One frame of point cloud data includes multiple point cloud pixels, each point cloud pixel corresponds to a receiving unit, and each point cloud pixel includes the ranging distance and reflectivity of the target object. Since each point cloud pixel is generated from data collected by multiple receiving devices, as long as a photon can be detected by any one of the multiple receiving devices, a valid point cloud pixel can be generated. This method can significantly improve the probability of capturing photons, thereby effectively improving the dynamic range of a single point cloud pixel. Attached Figure Description

[0015] One or more embodiments are illustrated by way of example with reference to the accompanying drawings, which are not intended to limit the embodiments, and elements having the same reference numerals in the drawings are designated as similar elements.

[0016] Figure 1 This is a schematic diagram of the lidar chip provided in the embodiments of this application. Figure 1 ;

[0017] Figure 2 This is a schematic diagram of the lidar chip provided in the embodiments of this application. Figure 2 ;

[0018] Figure 3 This is a schematic diagram of the lidar chip provided in the embodiments of this application. Figure 3 ;

[0019] Figure 4 This is a schematic diagram of the lidar chip provided in the embodiments of this application. Figure 4 ;

[0020] Figure 5 This is a schematic diagram of the lidar chip provided in the embodiments of this application. Figure 5 ;

[0021] Figure 6 This is a schematic diagram of the lidar chip testing system provided in an embodiment of this application;

[0022] Figure 7 This is a flowchart of the point cloud generation method provided in the embodiments of this application.

[0023] Figure label:

[0024] 1000, LiDAR chip testing system; 200, Light source; 100, LiDAR chip; 40, Testing module; 41, Observation and inspection unit; 42, First control unit; 43, Coordinate recording unit; 44, Second control unit; 45, Storage unit; 30, Point cloud data processing module; 31, Merging unit; 32, Histogram accumulation unit; 33, Point cloud generation unit; 34, Selection unit; 35, Expected value configuration unit; 20, Receiving module; 21, Receiving array; 22, Time conversion unit; 10, Transmitting module; T1, Target object. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and thoroughly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0026] It should be noted that when an element is described as "connected" to another element, it can be directly connected to the other element, or there can be one or more intermediate elements between them.

[0027] Furthermore, the technical features involved in the various embodiments of this application described below can be combined with each other as long as they do not conflict with each other.

[0028] Please refer to Figure 1 , Figure 1 This is a schematic diagram of a lidar chip provided in an embodiment of this application. Figure 1 As shown, the lidar chip 100 includes a transmitting module 10, a receiving module 20, and a point cloud data processing module 30.

[0029] The emitting module 10 is used to emit a probe laser. In some embodiments, the emitting module 10 includes an emitting array, which includes multiple emitting devices. In some embodiments, the emitting devices are vertical-cavity surface-emitting lasers (VCSELs) or edge-emitting lasers (EELs).

[0030] The receiving module 20 is used to receive the echo laser formed by the detection laser reflected by the target object T1. The receiving module 20 includes a receiving array 21 and a time conversion unit 22. The receiving array 21 includes multiple receiving units, namely a first receiving unit A1, a second receiving unit A2, ..., a Kth receiving unit AK, where K is an integer greater than 1. Each receiving unit includes multiple receiving devices; that is, the first receiving unit A1, the second receiving unit A2, ..., the Kth receiving unit AK each include multiple receiving devices, and the number of receiving devices included in different receiving units can be the same or different. In some embodiments, the receiving device is a single-photon avalanche diode (SPAD). When a single photon is incident on the SPAD, it can trigger avalanche breakdown, thereby generating an avalanche current. The avalanche current is processed by subsequent circuitry to finally output a detectable digital pulse.

[0031] The time conversion unit 22 is used to obtain the photon count value of each receiving device at different flight times based on the flight time of the photons received by each receiving device.

[0032] The time conversion unit 22 can determine the time from the emission of the probe laser from the transmitting module 10 to the avalanche breakdown of each receiving device; this time is the photon's flight time. The flight time is pre-divided into multiple time periods. Based on the flight time of the photon received by each receiving device, the photon is assigned to the corresponding time period, and the number of photons in each time period is counted, thereby obtaining the photon count value of each receiving device in different time periods, that is, obtaining the photon count value of each receiving device under different flight times.

[0033] The point cloud data processing module 30 is connected to the receiving module 20. The point cloud data processing module 30 is used to generate point cloud data based on the photon count value of each receiving device at different flight times. One frame of point cloud data includes multiple point cloud pixels. Each point cloud pixel corresponds to a receiving unit. Each point cloud pixel includes the ranging distance and reflectivity of the target object.

[0034] In related technologies, a point cloud pixel is usually generated from data collected by a receiving device. However, for a single receiving device, its photon detection efficiency is usually 30% to 40%, that is, when a photon hits a single-photon avalanche diode, there is only a 30% to 40% probability that it will be successfully recorded as a valid event, resulting in a low dynamic range of the obtained point cloud pixel.

[0035] In the embodiments of this application, a point cloud pixel is assigned to a receiving unit. That is, a point cloud pixel is generated from data collected by a receiving unit (i.e., multiple receiving devices). Therefore, as long as a photon can be detected by any one of the multiple receiving devices, a valid point cloud pixel can be generated. This method can significantly increase the probability of capturing photons, thereby effectively improving the dynamic range of a single point cloud pixel. For example, assuming the photon detection efficiency of a single receiving device is typically 30%, and a receiving unit (corresponding to one point cloud pixel) includes four receiving devices, the photon detection efficiency of the receiving unit is approximately 1. (1 0.3) 4 ≈76% > 30%, indicating that the dynamic range of a single point cloud pixel has been significantly improved.

[0036] In some embodiments, the receiving device is a single-photon avalanche diode. For example... Figure 2 As shown, the lidar chip 100 also includes a test module 40, and the point cloud data processing module 30 includes a merging unit 31, a histogram accumulation unit 32, and a point cloud generation unit 33.

[0037] The test module 40 is used to test the receiving array 21 to identify faulty receiving devices within it. Failed receiving devices include those in the receiving array 21 that cannot perform photoelectric conversion properly due to physical damage, manufacturing defects, aging, or environmental stress. Failed receiving devices can negatively impact the performance of the lidar chip 100, potentially leading to poor ranging accuracy.

[0038] When the transmitting module 10 emits a probe laser once, the merging unit 31 replaces the photon count value of the failed receiving device in each receiving unit with the expected value, and determines the photon count value of each receiving unit for the same flight time based on the photon count values ​​of all receiving devices in each receiving unit for the same flight time. Specifically, for a receiving unit (one point cloud pixel), it corresponds to multiple receiving devices. When there is a failed receiving device among these multiple receiving devices, the photon count value of the failed receiving device needs to be replaced with the expected value, while the photon count values ​​of the non-failed receiving devices among the multiple receiving devices remain unchanged. Then, the photon count value of the receiving unit is determined by combining the photon count value of the failed receiving device (which is now the expected value) and the photon count values ​​of the non-failed receiving devices. The expected value is a pre-set photon count value, which can be set according to different application scenarios.

[0039] The histogram accumulation unit 32 is used to accumulate multiple photon count values ​​of each receiving unit at different flight times to obtain an accumulated histogram. Each time a probe laser is emitted, a photon count value is obtained for each receiving unit. Specifically, the transmitting module 10 emits multiple probe lasers, and the histogram accumulation unit 32 superimposes the multiple photon count values ​​obtained from the multiple probe lasers emitted by the transmitting module 10 to obtain a statistically significant photon count histogram, which is the accumulated histogram.

[0040] The point cloud generation unit 33 is used to determine the ranging distance and reflectivity corresponding to each point cloud pixel based on the accumulated histogram, in order to generate point cloud data for the current frame. In some embodiments, the point cloud generation unit 33 obtains the ranging distance corresponding to each point cloud pixel based on the peak value of the photon counting histogram corresponding to each receiving unit. In other embodiments, the point cloud generation unit 33 can also obtain the reflectivity corresponding to each point cloud pixel based on the height and width of the photon counting histogram corresponding to each receiving unit.

[0041] In related technologies, when some receiving devices in a receiving array fail, the receiving array is usually discarded, which leads to a low yield rate for the receiving array.

[0042] In the embodiments of this application, by replacing the photon count value of the failed receiving device with the expected value, the receiving array can continue to be used without affecting the final point cloud quality, thus effectively improving the yield of the receiving array.

[0043] In some embodiments, such as Figure 3 As shown, the point cloud data processing module 30 also includes a selection unit 34, and the merging unit 31 includes multiple adders (not shown) and multiple path selectors (not shown).

[0044] In this system, each path selector corresponds one-to-one with a receiving device. The first input of each path selector is used to obtain the photon count value of the corresponding receiving device, and the second input of each path selector is used to obtain the expected value of the corresponding receiving device. A selection unit 34 is connected to the control terminal of each path selector and is used to select the second input and output of the path selector corresponding to a failed receiving device, or to select the first input and output of the path selector corresponding to a valid receiving device. An adder is used to accumulate the values ​​at the output terminals of the path selectors corresponding to the receiving devices in each receiving unit to obtain the photon count value of each receiving unit.

[0045] Figure 4 An exemplary embodiment shows that any receiving unit (such as the first receiving unit A1) includes four receiving devices, with photon count values ​​of DA1, DA2, DA3, and DA4, respectively. Correspondingly, four expected values ​​are set, namely DB1, DB2, DB3, and DB4. Correspondingly, the merging unit 31 includes four path selectors, namely MU1, MU2, MU3, and MU4. Correspondingly, when the selected adder has two inputs and one output, the merging unit 31 includes three adders, namely ADD1, ADD2, and ADD3.

[0046] Specifically, since the receiving unit includes four receiving devices, four path selectors are required. Path selector MU1 receives the photon count value DA1 at its first input and the expected value DB1 at its second input; path selector MU2 receives the photon count value DA2 at its first input and the expected value DB2 at its second input; path selector MU3 receives the photon count value DA3 at its first input and the expected value DB3 at its second input; path selector MU4 receives the photon count value DA4 at its first input and the expected value DB4 at its second input. Adder ADD1 receives the output of path selector MU1 at its first input and the output of path selector MU2 at its second input; adder ADD2 receives the output of path selector MU3 at its first input and the output of path selector MU4 at its second input; adder ADD3 receives the output of adder ADD1 at its first input and the output of adder ADD2 at its second input, and its output output is the photon count value of the point cloud pixel corresponding to the receiving unit.

[0047] Selection unit 34 is connected to the control terminals of path selectors MU1, MU2, MU3, and MU4. Selection unit 34 selects the second input and output terminals of the path selector corresponding to the failed receiving device in the receiving unit, or selects the first input and output terminals of the path selector corresponding to the unfailed receiving device in the receiving unit.

[0048] In a specific embodiment, the receiver with a photon count value of DA1 is a failed receiver, while the receivers with photon count values ​​of DA2, DA3, and DA4 are valid receivers. In this case, selection unit 34 selects the second input and output of path selector MU1, and selects the first input and output of path selectors MU2, MU3, and MU4. At this time, the first input of adder ADD1 receives the expected value DB1, and the second input receives the photon count value DA2; the first input of adder ADD2 receives the photon count value DA3, and the second input receives the photon count value DA4; the first input of adder ADD2 receives the result of superimposing the expected value DB1 and the photon count value DA2, and the second input receives the result of superimposing the photon count values ​​DA3 and DA4; the output of adder ADD3 outputs the result of superimposing the expected value DB1, the photon count value DA2, the photon count value DA3, and the photon count value DA4. The output of adder ADD3 is the photon count value of the point cloud pixel corresponding to the receiving unit.

[0049] It is understood that this embodiment is illustrated using a single receiving unit as an example, and the number of adders and path selectors also corresponds to one receiving unit. However, for the receiving array 21, which includes K receiving units, the number of adders and path selectors should correspond to the K receiving units. In a specific embodiment, each of the K receiving units includes 4 receiving devices, so the merging unit 31 may include 4K path selectors and 3K adders.

[0050] In some embodiments, such as Figure 5 As shown, the point cloud data processing module 30 also includes an expected value configuration unit 35.

[0051] The current frame point cloud is the Mth frame point cloud, which includes first point cloud pixels, each corresponding to a first receiving unit. The expected value configuration unit 35 is used to set an initial value as the expected value corresponding to a failed receiving device in the first receiving unit within the previous N frame point clouds, and to set the average photon count value of the first receiving units within the previous M-1 frame point clouds as the expected value corresponding to a failed receiving device in the first receiving unit within the Mth frame point cloud. The transmitting module obtains one frame point cloud each time it emits a probe laser, where M and N are integers greater than 1, and M is greater than N.

[0052] Specifically, the first point cloud pixel is any point cloud pixel in the current frame's point cloud. The initial value is a pre-set photon count value, which can be set based on the actual application scenario; this embodiment does not impose specific limitations on this.

[0053] For the first N frames of point cloud, since there is not enough historical frame data available, the pre-set photon count value (i.e., the initial value) can only be used as the expected value corresponding to the failed receiving device in the first receiving unit, so as to replace the photon count value of the failed receiving device in the first receiving unit with the expected value, that is, to replace the photon count value of the failed receiving device in the first point cloud pixel with the initial value.

[0054] After the Nth frame of the point cloud, since there is enough historical frame data available, the data from the previous M-1 frames of the point cloud can be used as a reference to adjust the Mth frame of the point cloud. Specifically, the average photon count value of the first receiving unit in the previous M-1 frames of the point cloud is used as the expected value corresponding to the failed receiving device in the first receiving unit of the Mth frame of the point cloud. The photon count value of the failed receiving device in the first receiving unit of the Mth frame of the point cloud is then replaced with the expected value, that is, the photon count value of the failed receiving device corresponding to the first point cloud pixel in the Mth frame of the point cloud is replaced with the average photon count value of the first point cloud pixel in the previous M-1 frames of the point cloud.

[0055] In a specific embodiment, N is 2, and M is an integer greater than or equal to 3. For the first two frames of point clouds, i.e., the first and second frames, the photon count values ​​of the failed receivers are replaced with the initial values. For the third frame and subsequent frames, the photon count values ​​of the failed receivers in the first receiving unit (first point cloud pixel) of the third frame are replaced with the average of the photon count values ​​of the first receiving unit in the first two frames; the photon count values ​​of the failed receivers in the first receiving unit of the fourth frame are replaced with the average of the photon count values ​​of the first receiving unit in the first three frames; and so on, until the photon count value of the failed receiver in the first receiving unit of the Mth frame is replaced with the average of the photon count values ​​of the first receiving unit in the first M-1 frames.

[0056] In this way, firstly, it can be ensured that the photon count value of the failed receiver can be corrected from the first frame, without waiting for historical accumulation. Secondly, using the average photon count value of the first receiver unit in the point cloud of the previous M-1 frames as a reference is equivalent to accumulating the signal in the time domain, so as to reconstruct the photon count value of the failed receiver using historical valid observations, thereby improving the reliability of the expected value. Furthermore, the expected value can be automatically adjusted according to changes in scene illumination, reflectivity, exposure time, etc., avoiding overcorrection or undercorrection caused by using fixed values.

[0057] Please refer to Figure 6 , Figure 6This is a schematic diagram of a lidar chip testing system provided in an embodiment of this application. Figure 6 As shown, the lidar chip testing system 1000 includes a lidar chip 100 and a light source 200 as described in any embodiment of this application.

[0058] Here, light source 200 refers to a light radiation source from the external environment of the lidar chip 100, such as sunlight.

[0059] The receiving module 20 in the lidar chip 100 also includes a quenching circuit 23. The quenching circuit 23 is connected to the receiving array 21 and is used to quench and restore the receiving array 21. Specifically, the quenching circuit 23 is used to immediately reduce the bias voltage below the breakdown voltage after the single-photon avalanche diode triggers avalanche breakdown, terminating the avalanche process, and restoring the single-photon avalanche diode within a safe time, enabling the single-photon avalanche diode to regain its detection capability, thereby achieving repeated and reliable detection of a single photon.

[0060] The test module 40 in the lidar chip 100 includes an observation and inspection unit 41, a first control unit 42, a coordinate recording unit 43, a second control unit 44, and a storage unit 45.

[0061] The system includes a first control unit 42 connected to the light source 200, which controls the light source 200 to emit light or stop emitting light. An observation and inspection unit 41 connected to the first control unit 42 and the quenching circuit 23 is used to identify receivers that do not exhibit an avalanche effect as failed receivers in a first test mode, and to identify receivers that exhibit an avalanche effect as failed receivers in a second test mode. The first control unit 42 controls the light source 200 to emit light in the first test mode, and controls the light source 200 to stop emitting light in the second test mode. A coordinate recording unit 43 connected to the observation and inspection unit 41 records the position of the failed receiver. A second control unit 44 connected to the coordinate recording unit 43, and a storage unit 45 connected to the second control unit 44 writes the position of the failed receiver to the storage unit 45.

[0062] Specifically, the receiving array 21 is first put into operation. The first control unit 42 controls the light source 200 to emit light to execute the first test mode. At this time, the light source 200 illuminates the receiving array 21, and the receiving array 21 is under strong light illumination. The receiving devices in the receiving array 21 receive photons, generating an avalanche effect. Then, the quenching circuit 23 quenches and restores the receiving array 21. Next, the observation and inspection unit 41 determines whether each receiving device has experienced an avalanche effect, identifies the receiving devices that have not experienced an avalanche effect, identifies the receiving devices that have not experienced an avalanche effect as failed receiving devices, and sends the position of the receiving devices that have not experienced an avalanche effect to the coordinate recording unit 43 for recording. The second control unit 44 is used to write the position of the failed receiving devices to the storage unit 45.

[0063] In a specific embodiment, the observation and inspection unit 41 determines whether each receiving device exhibits an avalanche effect and identifies receiving devices that do not exhibit an avalanche effect as follows: After repeatedly executing the first test mode to perform multiple tests (e.g., 3 times) on the receiving array 21, if the observation and inspection unit 41 determines that the first receiving device in the receiving array 21 has consistently not exhibited an avalanche effect, then the first receiving device is determined to be a receiving device without an avalanche effect, thus identifying the first receiving device as a failed receiving device. In this way, repeated testing eliminates randomness and improves the reliability of the determination.

[0064] Subsequently, the first control unit 42 controls the light source 200 to stop emitting light to execute the second test mode. At this time, the light source 200 stops illuminating the receiving array 21, and the receiving array 21 is in a dark environment. The receiving devices in the receiving array 21 will not receive photons, thus preventing the avalanche effect. Next, the observation and inspection unit 41 determines whether each receiving device exhibits an avalanche effect, identifies the receiving devices exhibiting an avalanche effect as failed receiving devices, and sends the location of the failed receiving devices to the coordinate recording unit 43 for recording. The second control unit 44 is used to write the location of the failed receiving devices to the storage unit 45.

[0065] In a specific embodiment, the observation and inspection unit 41 determines whether each receiving device exhibits an avalanche effect, and the specific process for identifying the receiving device exhibiting an avalanche effect is as follows: After repeatedly executing the second test mode to perform multiple tests (e.g., 3 times) on the receiving array 21, if the observation and inspection unit 41 determines that the second receiving device in the receiving array 21 consistently exhibits an avalanche effect, then the second receiving device is determined to be a receiving device exhibiting an avalanche effect, and thus identified as a failed receiving device. In this way, repeated testing eliminates randomness and improves the reliability of the determination.

[0066] Please refer to Figure 7 , Figure 7 A flowchart illustrating a point cloud generation method provided in an embodiment of this application. The point cloud generation method is applied to the LiDAR chip 100 in any embodiment of this application. Figure 7 As shown, the point cloud generation method includes the following steps S710 to S750.

[0067] Step S710: Test the receiving array to identify the faulty receiving devices in the receiving array.

[0068] In some embodiments, step S710 can be implemented through the following steps: placing the receiving array in a strong light environment and identifying the receiving devices that do not exhibit an avalanche effect as failed receiving devices; placing the receiving array in a dark environment and identifying the receiving devices that exhibit an avalanche effect as failed receiving devices.

[0069] Specifically, first, the receiving array 21 is put into operation. Then, the receiving array 21 is placed in a strong light irradiation environment. The receiving devices in the receiving array 21 receive photons, generating an avalanche effect. Then, the quenching circuit 23 is used to quench and restore the receiving array 21. Next, it is determined whether each receiving device has experienced an avalanche effect, and the receiving devices that have not experienced an avalanche effect are identified. The receiving devices that have not experienced an avalanche effect are then identified as failed receiving devices.

[0070] Next, the receiving array 21 is placed in a dark environment. The receiving devices in the receiving array 21 will not receive photons, thus preventing the avalanche effect. Then, it is determined whether each receiving device exhibits the avalanche effect, and the receiving devices exhibiting the avalanche effect are identified as failed receiving devices.

[0071] Step S720: Obtain the photon count value of each receiving device at different flight times.

[0072] Specifically, the time from when the probe laser is emitted by the transmitting module 10 to when each receiving device experiences avalanche breakdown is the photon's flight time. This flight time is pre-divided into multiple time periods. Based on the flight time of the photon received by each receiving device, the photon is assigned to the corresponding time period, and the number of photons in each time period is counted. This yields the photon count value for each receiving device at different time periods, i.e., the photon count value for each receiving device at different flight times.

[0073] Step S730: Replace the photon count value of the failed receiving device in each receiving unit with the expected value, and determine the photon count value of each receiving unit for the same flight time based on the photon count values ​​of all receiving devices in each receiving unit for the same flight time.

[0074] Specifically, for a receiving unit (point cloud pixel), it corresponds to multiple receiving devices. When there is a failed receiving device among these multiple receiving devices, the photon count value of the failed receiving device needs to be replaced with the expected value, while the photon count value of the non-failed receiving devices among the multiple receiving devices remains unchanged. Then, by combining the photon count value of the failed receiving device (which is the expected value at this time) and the photon count value of the non-failed receiving devices, the photon count value of the receiving unit is determined.

[0075] In some embodiments, before performing step S730, the point cloud generation method further includes the following steps: determining the location of the failed receiving device; and determining the receiving unit (point cloud pixel) corresponding to the failed receiving device according to the location of the failed receiving device and the receiving device mapping relationship lookup table.

[0076] The receiver device mapping lookup table defines a one-to-one correspondence between each receiver device and a receiver unit (point cloud pixel) in the receiver array 21. Therefore, by substituting the location of a failed receiver device into the receiver device mapping lookup table, the corresponding receiver unit (point cloud pixel) can be obtained. Through the mapping relationship between receiver devices and their corresponding receiver units, the receiver unit corresponding to a failed receiver device can be quickly identified, significantly reducing the false detection rate and false negative rate of receiver devices, which is beneficial for subsequent accurate correction of failed receiver devices.

[0077] Step S740: Accumulate the multiple photon count values ​​of each receiving unit at different flight times to obtain the accumulated histogram, wherein each time a probe laser is emitted, a photon count value is obtained for each receiving unit.

[0078] Specifically, the transmitting module 10 emits multiple probe lasers. The multiple photon count values ​​obtained from the multiple probe lasers emitted by the transmitting module 10 are superimposed to obtain a statistically significant photon count histogram, which is the accumulated histogram.

[0079] Step S750: Based on the accumulated histogram, determine the ranging distance and reflectivity corresponding to each point cloud pixel to generate the current frame point cloud data.

[0080] Specifically, in some embodiments, the point cloud generation unit 33 obtains the ranging distance corresponding to each point cloud pixel based on the peak value of the photon counting histogram corresponding to each receiving unit. In other embodiments, the point cloud generation unit 33 can also obtain the reflectivity corresponding to each point cloud pixel based on the height and width of the photon counting histogram corresponding to each receiving unit.

[0081] In related technologies, when some receiving devices in a receiving array fail, the receiving array is usually discarded, which leads to a low yield rate for the receiving array.

[0082] In the embodiments of this application, by replacing the photon count value of the failed receiving device with the expected value, the receiving array can continue to be used without affecting the point cloud quality, thus effectively improving the yield of the receiving array.

[0083] In some embodiments, the point cloud generation method further includes the following steps: the current frame point cloud is the Mth frame point cloud, the current frame point cloud includes a first point cloud pixel, and the first point cloud pixel corresponds to a first receiving unit; an initial value is used as the photon count value of the first receiving unit in the previous N frame point clouds, and the average of the photon count values ​​of the first receiving unit in the M-1 frame point clouds is used as the photon count value of the first receiving unit in the Mth frame point cloud; wherein, a frame point cloud is obtained each time a probe laser is emitted, M and N are integers greater than 1, and M is greater than N.

[0084] Specifically, for the first N frames of point cloud, since there is not enough historical frame data available, the pre-set photon count value (i.e., the initial value) can be used as the expected value corresponding to the failed receiving device in the first receiving unit, so as to replace the photon count value of the failed receiving device in the first receiving unit with the expected value, that is, to replace the photon count value of the failed receiving device in the first point cloud pixel with the initial value.

[0085] After the Nth frame point cloud, since there is enough historical frame data available, the data from the previous M-1 frames point clouds can be used as a reference to adjust the Mth frame point cloud. Specifically, the average photon count value of the first receiving unit in the previous M-1 frames point cloud is used as the expected value corresponding to the failed receiving device in the first receiving unit in the Mth frame point cloud. The photon count value of the failed receiving device in the first receiving unit in the Mth frame point cloud is then replaced with the expected value, that is, the photon count value of the failed receiving device in the first receiving unit in the Mth frame point cloud is replaced with the average photon count value of the first receiving unit in the previous M-1 frames point cloud.

[0086] In this way, firstly, it can be ensured that the photon count value of the failed receiver can be corrected from the first frame, without waiting for historical accumulation. Secondly, using the average photon count value of the first receiver unit in the point cloud of the previous M-1 frames as a reference is equivalent to accumulating the signal in the time domain, so as to reconstruct the photon count value of the failed receiver using historical valid observations, thereby improving the reliability of the expected value. Furthermore, the expected value can be automatically adjusted according to changes in scene illumination, reflectivity, exposure time, etc., avoiding overcorrection or undercorrection caused by using fixed values.

[0087] This application also provides a non-volatile computer-readable storage medium storing computer-executable instructions that are executed by one or more processors, for example, executing the instructions described above. Figure 7 The method and steps.

[0088] This application also provides a computer program product, including a computing program stored on a non-volatile computer-readable storage medium. The computer program includes program instructions that, when executed by a computer, cause the computer to perform the point cloud generation method in any of the above method embodiments, for example, to perform the method described above. Figure 7 The method and steps.

[0089] The above description is merely an embodiment of this application and does not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

[0090] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Within the framework of this application, the technical features of the above embodiments or different embodiments can also be combined, and the steps can be implemented in any order. Those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A lidar chip, characterized in that, include: The transmitting module is used to emit detection lasers; The receiving module is used to receive the echo laser formed by the detection laser reflected by the target object; The receiving module includes a receiving array and a time conversion unit. The receiving array includes multiple receiving units, and each receiving unit includes multiple receiving devices. The time conversion unit is used to obtain the photon count value of each receiving device at different flight times based on the flight time of the photons received by each receiving device. The receiving device is a single-photon avalanche diode. A test module is used to test the receiving array and identify failed receiving devices in the receiving array; and A point cloud data processing module, connected to the receiving module, is used to replace the photon count value of the failed receiving device in each receiving unit with the expected value, and generate point cloud data based on the photon count value of each receiving device at different flight times. One frame of point cloud data includes multiple point cloud pixels, each point cloud pixel corresponds to one receiving unit, and each point cloud pixel includes the ranging distance and reflectivity of the target object. The point cloud data processing module also includes an expected value configuration unit; The current frame point cloud is the Mth frame point cloud, and the current frame point cloud includes a first point cloud pixel, which corresponds to a first receiving unit; The expected value configuration unit is used to set the initial value as the expected value corresponding to the failed receiving device in the first receiving unit in the first N frame point cloud, and to set the average photon count value of the first receiving unit in the first M-1 frame point cloud as the expected value corresponding to the failed receiving device in the first receiving unit in the Mth frame point cloud. Each time the transmitting module transmits the detection laser, it obtains a frame of the point cloud, where M and N are integers greater than 1, and M is greater than N.

2. The lidar chip according to claim 1, characterized in that, The point cloud data processing module includes: The merging unit, when the transmitting module emits the detection laser once, is used to replace the photon count value of the failed receiving device in each receiving unit with the expected value, and to determine the photon count value of each receiving unit in the flight time based on the photon count values ​​of all receiving devices in each receiving unit in the same flight time; A histogram accumulation unit is used to accumulate multiple photon count values ​​of each receiving unit at different flight times to obtain an accumulated histogram, wherein each emission of the probe laser yields a photon count value for each receiving unit; and The point cloud generation unit is used to determine the ranging distance and reflectivity corresponding to each point cloud pixel based on the accumulated histogram, so as to generate the current frame point cloud data.

3. The lidar chip according to claim 2, characterized in that, The point cloud data processing module further includes a selection unit, and the merging unit includes multiple adders and multiple path selectors; The path selector corresponds one-to-one with the receiving device. The first input terminal of each path selector is used to obtain the photon count value of the corresponding receiving device, and the second input terminal of each path selector is used to obtain the expected value of the corresponding receiving device. The selection unit is connected to the control terminal of each of the path selectors and is used to select the second input terminal and output terminal of the path selector corresponding to the failed receiving device, or to select the first input terminal and output terminal of the path selector corresponding to the non-failed receiving device. The adder is used to accumulate the values ​​at the output terminals of the path selectors corresponding to the receiving devices in each receiving unit to obtain the photon count value of each receiving unit.

4. A lidar chip testing system, comprising a lidar chip as described in any one of claims 1-3, characterized in that, The testing system also includes: light source; The receiving module in the lidar chip also includes: A quenching circuit, connected to the receiving array, is used to quench and restore the receiving array; The testing module in the lidar chip includes: A first control unit is connected to the light source and is used to control the light source to emit light or stop emitting light. An observation and inspection unit, connected to the first control unit and the quenching circuit, is used to identify the receiving device that does not exhibit an avalanche effect as the failed receiving device in a first test mode, and to identify the receiving device that exhibits an avalanche effect as the failed receiving device in a second test mode. Wherein, the first control unit controls the light source to emit light in the first test mode, and the first control unit controls the light source to stop emitting light in the second test mode; A coordinate recording unit, connected to the observation and inspection unit, is used to record the position of the failed receiving device; The second control unit is connected to the coordinate recording unit, and the storage unit is connected to the second control unit. The second control unit is used to write the position of the failed receiving device to the storage unit.

5. A point cloud generation method, applied to a lidar chip as described in any one of claims 1-3, characterized in that, The method includes: The receiving array is tested to identify the failed receiving device in the receiving array; Obtain the photon count value of each receiving device at different flight times; The photon count value of the failed receiving device in each receiving unit is replaced with the expected value, and the photon count value of each receiving unit at the same flight time is determined based on the photon count values ​​of all receiving devices in each receiving unit at the same flight time. The multiple photon count values ​​of each receiving unit at different flight times are accumulated to obtain an accumulated histogram, wherein each time a probe laser is emitted, a photon count value of each receiving unit is obtained; Based on the accumulated histogram, the ranging distance and reflectivity corresponding to each point cloud pixel are determined to generate the current frame point cloud data.

6. The method according to claim 5, characterized in that, The step of testing the receiving array to identify the failed receiving device in the receiving array includes: The receiving array is placed in a strong light irradiation environment, and the receiving devices that do not exhibit an avalanche effect are identified as the failed receiving devices. By placing the receiving array in a dark environment, the receiving devices that exhibit an avalanche effect are identified as the failed receiving devices.

7. The method according to claim 6, characterized in that, Before replacing the photon count value of the failed receiving device in each of the receiving units with the expected value, the method further includes: Determine the location of the failed receiving device; The receiving unit corresponding to the failed receiving device is determined by looking up the location of the failed receiving device and the receiving device mapping relationship table.

8. The method according to claim 5, characterized in that, The method further includes: The current frame point cloud is the Mth frame point cloud, and the current frame point cloud includes a first point cloud pixel, which corresponds to a first receiving unit. The initial value is used as the expected value corresponding to the failed receiving device in the first receiving unit in the first N frame point cloud, and the average value of the photon count of the first receiving unit in the first M-1 frame point cloud is used as the expected value corresponding to the failed receiving device in the first receiving unit in the M frame point cloud. Each time the detection laser is emitted, a frame of the point cloud is obtained, where M and N are integers greater than 1, and M is greater than N.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed, implements the point cloud generation method as described in any one of claims 5-8.

Images