CN117630866A

CN117630866A - Laser radar, resource allocation method for laser radar, and computer-readable storage medium

Info

Publication number: CN117630866A
Application number: CN202210956429.2A
Authority: CN
Inventors: 陈杰; 向少卿
Original assignee: Hesai Technology Co Ltd
Current assignee: Hesai Technology Co Ltd
Priority date: 2022-08-10
Filing date: 2022-08-10
Publication date: 2024-03-01
Also published as: WO2024032278A1

Abstract

The present invention provides a laser radar including: a plurality of emission units configured to emit probe light beams; the detection units and the emission units form a plurality of detection channels, and the detection units are configured to receive echoes of detection light beams emitted by the emission units and reflected by obstacles; the processing units are connected with the detection units and are configured to process the echoes received by the detection units and generate point clouds; wherein the echo processing relationship of the detection unit and the processing unit can be dynamically adjusted. The invention comprehensively considers hardware overhead and algorithm realization, dynamically tracks the resolution and the change of the point cloud data rate brought by one or more areas by using as few measurement resource processes, and is particularly suitable for the requirements of engineering on the balance among cost, volume and efficiency.

Description

Laser radar, resource allocation method for laser radar, and computer-readable storage medium

Technical Field

The present disclosure relates to the field of photoelectric detection, and in particular, to a laser radar, a method for allocating resources of the laser radar, and a computer readable storage medium.

Background

The laser radar is a commonly used ranging sensor, has the advantages of long detection distance, high resolution, strong active interference resistance, small volume, light weight and the like, and is widely applied to the fields of intelligent robots, unmanned aerial vehicles, automatic driving and the like. As a three-dimensional measurement system, a lidar achieves three-dimensional measurement coverage of a Field of View (FOV) of a measurement through an acquired point cloud. A multi-channel lidar based on Time of Flight (ToF) is suitable for situations where a large field of view is required to be scanned and a high-density point cloud is acquired.

In a multi-channel parallel working radar, in order to save TOF measurement resources, a scheme is conventionally adopted: multiple non-parallel channels share the same measurement resource in a time sharing way, and channels working in parallel adopt different measurement resources. For clarity, fig. 1 shows a schematic diagram of echo signal processing channels of a tree MUX architecture, where channels using the same measurement or processing resources are generally defined as one group, each channel within the group operates in time, and one or more channels in different groups operate in parallel. For example, a group includes channels CH1-16, which operate in a time-sharing manner and share the same measurement resource, and a tree-like Multiplexer (MUX) is selected to enable 16 channels in a BANK to sequentially access signal processing channels, that is, measurement resources, according to the operation sequence of the 16 channels in the BANK through address mapping. The tree MUX is basically characterized in that: the sets of the transceiving channels covered by each echo processing resource are not crossed with each other, and the connection relation is fixed. .

However, there are some cases where echo processing resources at the receiving end need to be dynamically configured. For example, in the field of autopilot, the field of view of the lidar may further define a region of interest or region of interest (ROI, region of Interests), such as a target region along the path of travel. Furthermore, if the ROI region can be moved dynamically, or needs to be anchored dynamically, and further a more refined scan is performed, this means that the amount of echo data to be processed for the ROI region will be different when the ROI region is not selected as the ROI, and the echo processing resources of the corresponding receiving end need to be configured dynamically. Alternatively, adjustment of horizontal or vertical scanning parameters of the lidar, such as scanning frequency, may require adaptive deployment on echo processing resources.

However, if dynamic TOF measurement resource allocation is to be performed, the technical problem faced is: because the cost, the volume and the efficiency are required to be balanced in engineering, when some channels do not correspond to the ROI area, the echo data output by the detector is less, and a minimum value exists; however, when some channels correspond to the ROI area, the amount of echo data generated by the channels is greatly increased, and there is a maximum value. If the radar complete machine configures measurement resources for each channel according to the maximum value, the efficiency is high, but the cost and the volume are increased; if the measurement resources are configured for each channel at a minimum, the cost and volume would be low, but efficiency would be sacrificed.

The matters in the background section are only those known to the public inventor and do not, of course, represent prior art in the field.

Disclosure of Invention

When dynamically tracking one or more areas, the echo and the point cloud data amount of the area to be tracked are very different from those of other areas, and in particular, cost, volume and efficiency are required to be balanced in engineering, and how to reasonably allocate measurement resources in real time to cope with the resolution of different areas and the change problem of the point cloud data rate is required to be solved, so the invention provides a laser radar, which comprises:

a plurality of emission units configured to emit probe light beams;

a plurality of detection units configured to receive the echo reflected by the obstacle by the detection light beam emitted by the emission unit; and

the processing units are connected with the detection units and are configured to process the echoes received by the detection units and generate point clouds;

wherein the echo processing relationship of the detection unit and the processing unit can be dynamically adjusted.

According to one aspect of the invention, at least one of the probe units is connected to a plurality of processing units, and for the probe units connected to the plurality of processing units, the echo thereof can be shared for processing by the plurality of processing units.

According to one aspect of the invention, when the field of view of the lidar comprises a first region and a second region, and the amount of echo data in the first region is greater than the amount of echo data in the second region, at least one detection unit corresponding to the first region is connected to a plurality of processing units.

According to an aspect of the present invention, the first region may include a plurality of sub-regions, and at least one probe unit corresponding to each sub-region is connected to a plurality of processing units.

According to one aspect of the invention, the plurality of sub-regions do not overlap each other.

According to one aspect of the invention, each detection unit corresponding to the first area is connected to a plurality of processing units.

According to one aspect of the invention, each detection unit corresponding to said second area is connected to a processing unit.

According to one aspect of the invention, each processing unit is connected to a plurality of detection units.

According to an aspect of the present invention, the lidar further includes a plurality of multiplexing units, the plurality of detection channels are divided into a plurality of groups, the detection units of each group are connected to one processing unit through one multiplexing unit, and the detection units of the plurality of groups are connected to one processing unit through one multiplexing unit.

According to one aspect of the invention, a minimum number of processing units to which the detection units satisfying each scanning mode are connected is determined corresponding to a preset plurality of scanning modes according to an expected point cloud result.

According to one aspect of the invention, the processing unit connected to each detection unit is determined based on the echo data processing duration of each detection unit.

According to one aspect of the invention, a processing unit connected to each detection unit is determined from the rotational speed of the lidar and the point cloud resolution.

According to one aspect of the invention, the processing units connected to each detection unit are determined based on the physical distance between the detection unit and the processing unit.

The present invention also provides a resource allocation method of a laser radar, the laser radar including a plurality of transmitting units, a plurality of detecting units, and a plurality of processing units, the plurality of transmitting units being configured to transmit detecting light beams, the resource allocation method including:

the detection units receive echoes of the detection light beams sent by the sending units and reflected by obstacles;

distributing the echo to one or more processing units connected with the detection unit to process the echo and generate a point cloud;

According to an aspect of the present invention, the resource allocation method further includes: when the field of view of the laser radar comprises a first area and a second area, and the echo data volume in the first area is larger than that in the second area, distributing the echo in the first area to one or more processing units connected with the detection units corresponding to the first area according to the echo data volume.

According to an aspect of the present invention, the first region may include a plurality of sub-regions, and the resource allocation method further includes: and distributing the echo data in the plurality of subareas to one or more processing units connected with the detection units corresponding to the plurality of subareas according to the echo data quantity in the plurality of subareas.

According to an aspect of the present invention, the resource allocation method further includes: and distributing the echo data in the second area to a processing unit connected with the detection unit corresponding to the second area according to the echo data quantity in the second area.

According to an aspect of the present invention, the plurality of probe channels are divided into a plurality of groups, and the resource allocation method further includes: the echo data of each set of detection channels is distributed to one or more processing units connected to the set of detection channels according to the echo data volume of the set of detection channels.

According to an aspect of the present invention, the resource allocation method further includes: and presetting a plurality of mapping tables according to the point cloud distribution, wherein the mapping tables comprise a range of a first area and processing units connected with detection units corresponding to the first area, and distributing echo data in the first area to the processing units connected with the detection units corresponding to the first area according to the echo data quantity in the first area.

According to an aspect of the present invention, the resource allocation method further includes: and switching the matched mapping table according to the change of the point cloud distribution.

According to an aspect of the present invention, the resource allocation method further includes: the mapping table is switched between the two probes.

The present invention also provides a computer readable storage medium comprising computer executable instructions stored thereon which, when executed by a processor, implement a resource allocation method as described above.

The invention comprehensively considers hardware overhead and algorithm realization, dynamically tracks the resolution and the change of the point cloud data rate brought by one or more areas by using as few measurement resource processes, and is particularly suitable for the requirements of engineering on the balance among cost, volume and efficiency.

Drawings

The accompanying drawings, which are included to provide a further understanding of the disclosure, illustrate and explain the exemplary embodiments of the disclosure and together with the description serve to explain the disclosure, and do not constitute an undue limitation on the disclosure. In the drawings:

FIG. 1 shows a schematic diagram of echo signal processing channels of a tree MUX architecture;

FIG. 2a shows a schematic view of a lidar vertical field of view according to an embodiment of the invention;

FIG. 2b shows a schematic view of a lidar horizontal field of view according to an embodiment of the invention;

FIGS. 3a and 3b show lidar diagrams of an embodiment of the invention;

FIG. 3c shows a lidar schematic diagram of another embodiment of the invention;

FIGS. 4a-4c are schematic diagrams illustrating the connection of a detection unit to a processing unit in accordance with various embodiments of the present invention;

FIG. 5a shows a schematic diagram of the division of a first region and a second region according to an embodiment of the present invention;

FIG. 5b shows a schematic view of a first region comprising a plurality of sub-regions according to one embodiment of the invention;

FIG. 5c shows a schematic view of a first region comprising a plurality of separate sub-regions according to one embodiment of the invention;

FIG. 5d shows a schematic view of a first region comprising a plurality of separate sub-regions according to another embodiment of the invention;

FIGS. 6a-6c are schematic diagrams illustrating scan patterns according to one embodiment of the invention;

FIG. 7 shows a schematic diagram of a scan pattern of another embodiment of the invention;

FIG. 8 illustrates a schematic view of a first zone walk of one embodiment of the present invention;

FIGS. 9a-9c are schematic diagrams illustrating the wired relationship of the BANK of FIG. 9 to a processing unit in accordance with various embodiments of the present invention;

fig. 10a shows a schematic diagram of echo data allocation under a scan pattern index=1 line according to an embodiment of the present invention;

fig. 10b shows a schematic diagram of echo data allocation under a scan pattern index=8 line according to an embodiment of the present invention.

Detailed Description

Hereinafter, only certain exemplary embodiments are briefly described. As will be recognized by those of skill in the pertinent art, the described embodiments may be modified in various different ways without departing from the spirit or scope of the present invention. Accordingly, the drawings and description are to be regarded as illustrative in nature and not as restrictive.

In the description of the present invention, it should be understood that the terms "center," "longitudinal," "transverse," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," etc. indicate or are based on the orientation or positional relationship shown in the drawings, merely for convenience of description and to simplify the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the invention. Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more of the described features. In the description of the present invention, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.

In the description of the present invention, it should be noted that, unless explicitly specified and limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be fixedly connected, detachably connected, or integrally connected, and may be mechanically connected, electrically connected, or may communicate with each other, for example; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements or interaction relationship between the two elements. The specific meaning of the above terms in the present invention can be understood by those of ordinary skill in the art according to the specific circumstances.

In the present invention, unless expressly stated or limited otherwise, a first feature "above" or "below" a second feature may include both the first and second features being in direct contact, as well as the first and second features not being in direct contact but being in contact with each other through additional features therebetween. Moreover, the first feature being "above," "over" and "on" the second feature includes the first feature being directly above and obliquely above the second feature, or simply indicating that the first feature is higher in level than the second feature. The first feature being "under", "below" and "beneath" the second feature includes the first feature being directly above and obliquely above the second feature, or simply indicating that the first feature is level less than the second feature.

The following disclosure provides many different embodiments, or examples, for implementing different structures of the invention. In order to simplify the present disclosure, components and arrangements of specific examples are described below. However, they are merely examples and are not intended to limit the present invention. Furthermore, the present invention may repeat reference numerals and/or letters in the various examples, which are for the purpose of brevity and clarity, and which do not themselves indicate the relationship between the various embodiments and/or arrangements discussed. In addition, the present invention provides examples of various specific processes and materials, but one of ordinary skill in the art will recognize the application of other processes and/or the use of other materials.

Fig. 2a shows a schematic view of a vertical field of view of a laser radar according to an embodiment of the present invention, where multiple lasers of the laser radar emit light according to a certain time sequence, and after the emitted light is collimated by a transmitting lens (group), the lasers are respectively directed in different directions when exiting from the laser radar. In a laser radar in which a laser and a detector form a detection channel, each channel is responsible for scanning within a certain vertical angle range, and the vertical field of view is the total angle range in which the laser radar can detect in the vertical direction, and the sum of the vertical angle ranges corresponding to all the channels together form the vertical field of view of the radar. In fig. 2a, n channels/beams are shown, n can be 16 or 32 or 40 or 64 or 128 or more or less, the vertical field of view is (x+y) °, covering-x to +y °, and is composed of the vertical angles of all lasers together. Wherein the vertical angle of the uppermost laser is +y° (+ denotes upwards with respect to the horizontal plane), i.e. is responsible for distance detection in the y ° direction, and the vertical angle of the lowermost laser is-x°, i.e. is responsible for distance detection in the downwards x ° direction. In a specific implementation, the vertical field of view of the actual lidar may be 40 °, may be 100 °, or may be another value.

Fig. 2b shows a schematic view of a horizontal field of view of a laser radar according to an embodiment of the present invention, in which the laser radar is shown in a cross-sectional view along a horizontal direction, which is generally perpendicular to a rotation axis of the radar, and the radar is driven by a rotating member such as a motor, or by turning a mirror, vibrating a mirror, MEMS, or liquid crystal, etc., so as to complete scanning of the entire horizontal field of view. The horizontal angle of view is the angle range that the lidar can detect in the horizontal direction, for example 360 ° for one rotation of the mechanical lidar, then 360 °. In the point cloud image output by the laser radar, the included angle between two adjacent detection points on the horizontal plane perpendicular to the rotating shaft is the horizontal angle resolution. In fig. 2b the cross section of the radar is circular, the invention is not limited thereto, and the cross section of the radar may also be other shapes, for example rectangular. The rotation shaft (or the rotation member) may be disposed at the center of the radar or may be disposed at a position on the left or right, which is within the scope of the present invention.

The vertical field of view and the horizontal field of view together comprise the field of view range of the lidar. The detection method is not only suitable for mechanical laser radar, but also suitable for solid laser radar and semi-solid laser radar. For example, solid-state lidar has a horizontal field of view typically less than 360 degrees, e.g., 120 degrees, and includes multiple lasers and multiple detectors forming multiple detection channels.

The present invention provides a laser radar including: a plurality of emission units configured to emit probe light beams; the detection units and the emission units form a plurality of detection channels, and the detection units are configured to receive echoes of detection light beams emitted by the emission units through the reflection of obstacles; the processing units are connected with the detection units and are configured to process the echoes received by the detection units and generate point clouds; wherein the echo processing relationship of the detection unit and the processing unit can be dynamically adjusted. At least one of the detecting units is connected with the plurality of processing units, and for the detecting units connected with the plurality of processing units, echo data of the detecting units can be shared and processed by the plurality of processing units. The invention comprehensively considers hardware overhead and algorithm realization, dynamically tracks the resolution and the change of the point cloud data rate brought by one or more areas by using as few measurement resource processes, and is particularly suitable for the requirements of engineering on the balance among cost, volume and efficiency.

The preferred embodiments of the present invention will be described below with reference to the accompanying drawings, it being understood that the preferred embodiments described herein are for illustration and explanation of the present invention only, and are not intended to limit the present invention.

Fig. 3a and 3b show a schematic view of a lidar according to an embodiment of the present invention, where the lidar 10 comprises a plurality of transmitting units 11, a plurality of detecting units 12 and a plurality of processing units 13 (not shown in fig. 3a, arranged on the back of the RX board, opposite to the detecting units), and specifically comprises:

the plurality of transmitting units 11 are configured to transmit probe beams, the plurality of detecting units 12 and the plurality of transmitting units 11 form a plurality of detecting channels, each detecting channel corresponds to a vertical direction, and the plurality of detecting units 12 are configured to receive echo data of the probe beams transmitted by the transmitting units 11 located in the same detecting channel and reflected by an obstacle. For example, the plurality of detection units 12 and the plurality of emission units 11 constitute a plurality of detection channels, each detection channel including at least one emission unit 11 and at least one detection unit 12. The plurality of detection channels can emit light in parallel or be divided into a plurality of groups of BANKs, and the emission units 11 in each BANK emit light in a non-parallel manner, and only one emission unit 11 in each BANK emits light at the same time. The sequence of the emission unit 11 can also be controlled by a time sequence, which is within the scope of the invention.

A plurality of processing units 13 connected to the detection unit 12 and configured to process the echo data and generate a point cloud.

Wherein the emitting unit comprises a laser, and the specific type can adopt VCSEL or EEL. The detection unit comprises an optoelectronic device for converting a received optical signal into an electrical signal, and specifically, for example, an APD, siPM or span array can be used. The processing resource can perform subsequent processing on the electric signal output by the photoelectric device, such as analyzing and obtaining the arrival time of the echo, the power of the echo and the like, so as to obtain the distance and/or reflectivity information of the obstacle; or for the emitting end, the processing resource may be a driving circuit for driving the laser to emit light. Depending on the type of optoelectronic device used, different processing units may be chosen. For example, for devices that will output analog signals, such as APD and SiPM, ADC may be used for further processing of subsequent signals; for devices such as SPAD arrays that output digital signals, a TDC may be used as the processing unit. In addition, the processing unit herein is a device having an upper limit of echo data amount/laser drive momentum in a period of time, that is, the processing unit has limited processing resources in a period of time, and the processable echo amount/drive amount in a unit time is the processing resources. The processing unit may allocate the total processing resources owned to driving different lasers or processing echo data of different detectors over a period of time. The upper limits of the processing resources of the respective processing units may be the same or different. For ease of description, the description herein below will default to a positive correlation of the number of processing units with the number of processing resources, i.e., allocating more processing units is equal to allocating more processing resources. In addition, regarding the processing resources, the receiving end is taken as an example more herein, and the actual transmitting end is the same, so that the detailed description is not repeated.

In order to support a scene where the laser radar needs to dynamically adjust scanning parameters, the echo processing relationship between the detection unit and the processing unit may be set to be dynamically adjustable. Specifically, at least one of the detecting units 12 may be connected to the plurality of processing units 13, and then, for the detecting unit 12 to which the plurality of processing units 13 are connected, data obtained after echo detection thereof may be shared and processed by the plurality of processing units 13, and data processing tasks may be dynamically allocated to one or more of the plurality of processing units 13 to be performed, thereby meeting the requirements of data processing instantaneity and data processing load.

In a specific embodiment, referring to fig. 4a, the lidar 10 comprises four detection units 12, namely detection unit 12A, detection unit 12B, detection unit 12C and detection unit 12D. The lidar 10 further comprises three processing units 13, namely a processing unit 13A, a processing unit 13B and a processing unit 13C. For example, because of the high detection resolution requirement, the detection unit 12A has a very large volume of echo data during a certain period of time (e.g. for a region of interest ROI, which is scanned more densely by the detection unit 12 and the corresponding laser, and thus has a large volume of echo data), whereas during this period of time these point cloud data cannot be processed in time by means of the processing unit 13A or 13B alone, and because it is possible that the processing unit 13A also needs to process echoes from other detectors 12A-only part of the resources can be used for processing echoes from the detection unit 12A. Thus, in the embodiment of fig. 4a, the detection unit 12A is connected to the processing unit 13A and the processing unit 13B, so that echo data received by the detection unit 12A can be shared by the two processing units 13A and 13B, so as to ensure that all processing units are not overloaded, and that all data of detectors, such as 12A to 12A, 12B, 12D and others not shown, in the lidar can be processed during this period. Furthermore, the echo data of the detecting unit 12B, the detecting unit 12C and the detecting unit 12D are relatively small in frequency, and the total amount of the point cloud data can be borne by one processing unit 13, and the three are respectively connected with the processing unit 13C, and the echo data received by the three are independently processed by the processing unit 13. As can be seen from fig. 4a, the four detection units 12 are connected with the three processing units 13, compared with the scheme that one detection unit 12 is connected with one processing unit 13, one processing unit 13 is saved, and the TOF measurement resources are reasonably allocated in real time to accept all echo data; compared with the scheme that one processing unit 13 is connected with a plurality of detection units 12, the method and the device solve the problem that the processing unit 13 cannot process echo data in time, and achieve efficient configuration of TOF measurement resources.

In addition, in the embodiment of fig. 4a, the detection unit 12A is connected to the processing unit 13A and the processing unit 13B, and the echo generated by the detection unit 12A may be processed by the processing unit 13A alone, may be processed by the processing unit 13B alone, or may be shared by the processing unit 13A and the processing unit 13B (for example, may be in proportion). The allocation of a particular data processing task may depend on the amount of echo data that needs to be processed, as well as the current task load of each processing unit 13A and 13B. For example, when the amount of echo data currently generated by the detection unit 12A is small (smaller than a preset value), the echo data generated by the detection unit 12A may be processed by the processing unit 13A or the processing unit 13B alone (may be selected according to a preset order, for example); when the amount of echo data currently generated by the detection unit 12A is large (larger than a preset value), the allocation can be dynamically performed between the processing unit 13A and the processing unit 13B according to a preset ratio, thereby ensuring the real-time performance of echo processing.

It should be noted that, the measurement resource, for the transmitting end, is the driving circuit channel of the laser; for the receiving end, it is the resource of echo processing, such as ADC or TDC channels, i.e. channels that assist TOF calculation. In addition, the echo data amount of the ROI area and the non-ROI area are greatly different, the horizontal resolution of the ROI area may be x times that of the non-ROI area, the vertical resolution may be y times that of the non-ROI area, and the whole is xy times the point cloud data.

In another specific embodiment, referring to fig. 4B, the lidar 10 comprises four detection units 12, namely detection unit 12A, detection unit 12B, detection unit 12C and detection unit 12D. The laser radar 10 further comprises four processing units 13, namely a processing unit 13A, a processing unit 13B, a processing unit 13C and a processing unit 13D. For example, the echo data amount of the detecting unit 12A is often very large, which is connected to the processing unit 13A and the processing unit 13B, and the echo data received by the detecting unit 12A can be shared by the two processing units 13. Further, the echo data amount of the detecting unit 12B is also relatively large, but one processing unit 13 is enough to bear, and the detecting unit 12B is connected with the processing unit 13C, and the processing unit is solely shared; the echo data amounts of the detecting unit 12C and the detecting unit 12D are relatively small constantly, and the total amount of the point cloud data can be borne by one processing unit 13, so that the two are respectively connected with the processing unit 13D, and the echo data received by the two are independently processed by the processing unit 13D. As can be seen from fig. 4b, the four detecting units 12 are connected with the four processing units 13, and compared with the scheme that one detecting unit 12 is connected with one processing unit 13 or one processing unit 13 is connected with a plurality of detecting units 12, the reasonable and efficient configuration of measuring resources is realized.

In another specific embodiment, referring to fig. 4C, the lidar 10 includes five detection units 12, namely detection unit 12A, detection unit 12B, detection unit 12C, detection unit 12D, and detection unit 12E. The lidar 10 further comprises five processing units 13, namely a processing unit 13A, a processing unit 13B, a processing unit 13C, a processing unit 13D and a processing unit 13E. For example, the echo data amount of the detecting unit 12A is very large regularly, which is connected to the processing unit 13A and the processing unit 13B, and the echo data received by the detecting unit 12A can be dynamically shared by the two processing units 13. Further, the echo data of the detecting unit 12B and the echo data of the detecting unit 12C are equivalent and relatively large, the detecting unit 12B is connected with the processing units 13C and 13D, and the detecting unit 12C is connected with the processing units 13C and 13D, so that not only can the echo data of one detecting unit 12 be dynamically shared by the two processing units 13, but also the assistance of the two detecting units 12 by the one processing unit 13 can be realized. The echo data amounts of the detecting unit 12D and the detecting unit 12E are relatively small constantly, and the total amount of the point cloud data can be borne by one processing unit 13, so that the two are respectively connected with the processing unit 13E, and the echo data received by the two are independently processed by the processing unit 13. As can be seen from fig. 4c, the five detection units 12 are connected with the five processing units 13, and compared with the scheme that one detection unit 12 is connected with one processing unit 13 or one processing unit 13 is connected with a plurality of detection units 12, the reasonable and efficient configuration of TOF measurement resources is realized.

The above description has been made on the feature that "at least one of the probe units 12 is connected to the plurality of processing units 13, and the echo data of the probe unit 12 connected to the plurality of processing units 13 can be shared by the plurality of processing units 13" by the embodiment, which has the common feature that the probe unit 12A is connected to two processing units 13, the echo data received by the probe unit is shared by the two processing units 13, and further, the processing units 13 are allocated to other probe units 12 according to the echo data amount, so that the efficient configuration of measurement resources is realized, and the problems that one probe unit 12 is connected to one processing unit 13, the resources of some processing units 13 are insufficient, and the resources of some processing units 13 are excessive are avoided. It will be appreciated by those skilled in the art that the numbers in the above embodiments are merely exemplary, and the connection relation between the probe unit and the processing unit and the resources processed according to the echo data amount allocation are exemplary, and do not limit the present invention.

The system architecture and resource allocation strategy of lidar 10 were described macroscopically above. The echo data amounts for different regions are very different, and the need for fast dynamic tracking and thus finer scanning of the region of interest requires optimization of measurement resources, especially the processing unit 13 at the receiving end needs to be dynamically configured. The hardware wiring between the plurality of processing units 13 and the plurality of detection units 12 and the dynamic allocation of echo data are therefore the focus of the present invention. In the present invention, a "processing unit" is to be understood broadly to include both hardware resources and data computing and processing capabilities that each hardware resource may provide in time-sharing or in parallel. For example, the plurality of processing units depicted in fig. 4a-4c may comprise a plurality of data processing devices, such as hardware devices, e.g., TDCs and/or ADCs, and may also comprise resources of data computing and processing capabilities that a data processing device is capable of providing. The former is easy to understand, and the latter is described with emphasis below. For a data processing device, its data processing capacity can be divided into different time slices (time slots), which can be allocated to different probe units 12, i.e. for processing echo data generated by different probe units 12. In this case, one time slice of one data processing apparatus may correspond to one processing unit shown in fig. 4a to 4c, and a plurality of time slices of one or more data processing apparatuses may correspond to a plurality of processing units 13 shown in fig. 4a to 4 c. Instead of time slices, the data processing device may also run different threads simultaneously, each for processing echo data generated by a different detection unit 12. In this case one thread of one data processing device corresponds to one processing unit shown in fig. 4a-4c, and multiple threads of one or more data processing devices correspond to multiple processing units 13 shown in fig. 4a-4 c. In addition, the time slices or threads of the data processing apparatus may be synchronized, i.e. may be used simultaneously for processing echo data generated by different detection units 12. Accordingly, the term "processing unit" in the present invention is understood to have a broad meaning.

In order to effectively utilize the processing units 13, a concept of a resource pool is introduced, namely, all the processing units 13 are planned as a total resource pool, specifically, the hardware and software processing capabilities (such as time slices and threads) of all the processing units 13 are planned as a resource pool, wherein a plurality of processing units 13 can be connected with the same detecting unit 12 in an abutting mode, and thus the plurality of processing units 13 can share echo data of the detecting unit 12. The processing unit referred to in this specification may refer to an individual in a physical sense or may refer to an intangible processing resource, and is not intended to limit the present invention.

The above embodiment, although exemplarily describing the connection relationship between the probe units 12 and the processing unit 13, can cope with the case where the echo data amount received by each probe unit 12 is known in advance and the echo data amount is relatively fixed, and for the case of dynamic tracking and configuration according to the echo data amount, further refinement of the connection and resource allocation scheme is required. Before describing the scheme, the point cloud mode and the region division of the laser radar field of view are described, and the hardware connection relation between the detection unit and the processing unit is described by combining the region division.

According to a preferred embodiment of the invention, when the field of view of the lidar comprises a first region and a second region, and the amount of echo data in the first region is larger than the amount of echo data in the second region, at least one detection unit corresponding to the first region is connected to the plurality of processing units.

Fig. 5a shows a schematic diagram of the division of a first area and a second area according to an embodiment of the present invention, wherein the largest rectangular frame is the field of view of the lidar 10, the horizontal field of view, and the vertical field of view, and one area is selected as the first area and the other area is selected as the second area. The first area may be set according to an area where the radar is applied to an area of frequent interest or an area where an obstacle of interest is located in an ADAS scene, such as directly in front of a vehicle. It is obvious that the echo data volume of the first area will be larger than the echo data volume of the second area, and that more processing units 13 should be allocated to the detection units 12 corresponding to the first area, at least satisfying that at least one detection unit corresponding to the first area is connected to a plurality of processing units.

In practice, however, the obstacle of real interest occupies only a small area, and if the first area is too large, the processing units 13 allocated thereto may be too many, and a problem of uneven distribution of the processing resources may inevitably occur.

According to a preferred embodiment of the invention, the first region may comprise a plurality of sub-regions, with at least one detection unit corresponding to each sub-region being connected to a plurality of processing units.

Fig. 5b shows a schematic diagram of the first area according to an embodiment of the invention comprising a plurality of sub-areas, the first area being further divided into a plurality of sub-areas, e.g. sub-areas 1-4, only one or more sub-areas being activated per detection, by activated it being meant that in the activated sub-areas the amount of echo data (or the density of echo reception or the total number of echoes over a period of time) is higher than in the non-activated sub-areas due to an increased number of beams scanned and/or an increased frequency of scanning. For example, if an obstacle appears in the sub-area 1, the echo data amount of the sub-area 1 is relatively high, if the user chooses to dynamically track the obstacle, the scanning harness or scanning frequency in the sub-area 1 is activated, so that a point cloud with higher resolution is obtained for the obstacle selected by the user, and thus the generated echo data amount is relatively large, and the processable echo data amount per processing unit is limited in unit time, so that more processing units 13 are allocated to the corresponding detection unit 12 of the sub-area 1 (that is, the echo generated by the detection unit 12 is allocated to a plurality of processing units 13 connected with the processing unit to be processed, or one of the processing units 13 is allocated to only process the data of the detection unit 12 in a next period of time). While the echo data volume of the sub-areas 2-4 and the second area is relatively small, the processing units required for each respective detection unit 12 are also small (thus, for example, one detection unit 12 in the sub-area 2-4, even though it may be connected to a plurality of processing units, can be allocated to only one of the connected processing units 13 for processing due to the small echo data volume). It should be noted that, the point cloud data of a sub-area is high, which means that the resolution of the sub-area in the transverse direction or the longitudinal direction is improved, if the resolution is improved in the transverse direction, possibly by improving the transmitting frequency, the receiving frequency of the corresponding detector needs to be improved, which means that the echo data received by the detector in a certain period of time is more; if the longitudinal improvement is made, 1 echo is likely to be received in a period t1 before a certain detector, and 1 echo is currently received in a period t2, and t1 > t2, and echo data received by the detector in that unit time is higher.

According to a preferred embodiment of the invention, wherein the plurality of sub-areas do not overlap each other, the plurality of sub-areas may be detected simultaneously.

Fig. 5c shows a schematic view of a first region of an embodiment of the invention comprising a plurality of separate sub-regions, differing from the embodiment of fig. 5b in that: wherein the first region is divided into a plurality of separate sub-regions. Fig. 5c is only an exemplary illustration, and the present embodiment does not limit the size, coverage, number of sub-areas or the relative position between sub-areas. For example, during the running of the vehicle, if an obstacle appears in the sub-area 1, the echo data amount of the sub-area 1 is relatively high, if the user selects the region of interest to dynamically track the obstacle, the scanning harness or the scanning frequency in the sub-area 1 is activated, so that a point cloud with higher resolution is obtained for the obstacle selected by the user, and thus the generated echo data amount is relatively large, and the echo data amount that can be processed by each processing unit in a unit time is limited, so that more processing units 13 are allocated to the corresponding detection unit 12 of the sub-area 1 (that is, the echo generated by the detection unit 12 is allocated to a plurality of processing units 13 connected with the processing unit to be processed, or one of the processing units 13 is allocated to only process the data of the detection unit 12 in a subsequent period of time). While the echo data volume of the sub-area 4 and the second area is relatively small, so is the need for fewer processing units per detection unit 12.

Fig. 5d shows a schematic view of another first region of the invention comprising a plurality of separate sub-regions, differing from the embodiment of fig. 5c in that: wherein the first region is divided into a plurality of separate and differently sized sub-regions. Fig. 5d is only an exemplary illustration, and the present embodiment does not limit the size, coverage, number of sub-areas or the relative positions between sub-areas. For example, during the running of the vehicle, if an obstacle appears in the sub-area 4, the echo data amount of the sub-area 4 is relatively high, if the user selects the region of interest to dynamically track the obstacle, the scanning line beam or scanning frequency in the sub-area 4 is activated, so that a higher resolution point cloud is obtained for the obstacle selected by the user, and thus the generated echo data amount is relatively large, and each processing unit has a limited processable echo data amount in a unit time, so that more processing units 13 are allocated to the corresponding detection unit 12 of the sub-area 4 (that is, the echo generated by the detection unit 12 is allocated to a plurality of processing units 13 connected thereto for processing, or one of the processing units 13 is allocated to exclusively process the data of the detection unit 12 in a subsequent period of time). While the echo data volume of the sub-area 1 and the second area in this period of time is relatively small, so is the processing unit required for each corresponding detection unit 12.

The embodiment is used for dividing the first area into a plurality of subareas, so that the calculation force for positioning the area to be tracked according to the selected area of interest can be saved, and the system power consumption can be reduced. However, when switching between the sub-areas, a plurality of scanning modes may be preset, so as to further save calculation force and improve efficiency. As described further below.

Synchronous detection means that the laser radar can finish the detection of a plurality of mutually non-overlapped subareas in one detection process without switching or changing the field of view.

Some galvanometer-type lidars claim to be capable of tracking multiple areas in the following specific modes: the FOV of the field of view of the galvanometer is reduced to match the size of the sub-area corresponding to the obstacle, for example, the FOV of the galvanometer scan itself is 25 DEG horizontally+30 DEG vertically, the FOV is reduced to 5 DEG horizontally+6 DEG vertically for scanning the sub-area, and then the FOV is scanned only in the reduced FOV. The result is that: 1. each scan can only pay attention to 1 obstacle, only scan the subarea corresponding to the obstacle, if there are 2 or even more obstacles, then multiple scans need to be executed; 2. while scanning the sub-region corresponding to the obstacle, other regions are not scanned because the FOV has been reduced to exclude other regions from the scan. Therefore, the invention is at least different from the galvanometer type laser radar in that: multiple sub-areas which are not overlapped with each other can be arranged, and the multiple sub-areas can be synchronously detected in parallel, and the definition of the synchronization is that the field of view is not required to be switched, so that after one sub-area is prevented from being scanned, the sub-area is switched to be aligned with the other sub-area, the sub-area is scanned, and if an object appearing in the time difference between the two scans moves, an accurate detection result cannot be obtained.

The above embodiment achieves further refinement of the first region by providing sub-regions, i.e. regions requiring more processing units and regions requiring less processing units are refined, thereby more accurately matching the user selected obstacle or region of interest. In practice, however, the obstacle of interest is moving, or moving relative to the lidar, in order to track the obstacle of interest, a first area that is moved within the field of view of the lidar may be set, and the size of the first area may be adjustable, so as to achieve both area refinement and dynamic configuration of measurement resources.

According to a preferred embodiment of the invention, a plurality of scanning modes are preset according to one or more of the number, the size, the moving direction and the moving step length of the sub-areas in the first area, and different scanning modes correspondingly activate different sub-areas.

Fig. 6a-6c show schematic diagrams of scan patterns according to an embodiment of the invention, the largest rectangular box representing the full field of view achievable by a single scan, the horizontal field of view being horizontal and the vertical field of view, wherein the dashed box is the first area, to which more processing units 13 should be allocated for the detection units 12 corresponding; the relatively smaller solid rectangular box is a single activated sub-area, that is, a sub-area that can be actually activated in one scan, and once activated, more processing resources are allocated to the detection units 12 corresponding to the sub-area, where the processing resources may be more processing units 13, or may be that in a period of time, 1 or more processing units 13 are allocated to only process echo data from the detection units; the area between the largest rectangular box and the dotted rectangular box is the second area, and the echo data amount of this area is smaller, so that fewer processing units 13 can be allocated to the corresponding detection units 12.

For example, in the advanced driving assistance system ADAS application scenario, the area in front of is most focused on, as the first area shown by the dotted rectangle, the present embodiment subdivides the area into multiple activatable sub-areas, such as the single activated sub-area 1 shown in fig. 6a, the single activated sub-area 2 shown in fig. 6b, and the single activated sub-area 3 shown in fig. 6 c. Preferably, the vertical fields of view are arranged in sequence from above to below in the direction of movement. For example, the scanning pattern may be matched as the selected region of interest is closer to the vehicle during its travel, which in turn partially overlaps with the single activated sub-regions 1-3. The greater the number of single activated sub-regions, the smaller the field of view each sub-region covers, the smaller the step size of the movement, the better the fine match with the selected region of interest. The moving direction of the subareas can be set according to practical data, and the sizes, the coverage areas, the moving directions and the moving step sizes of the subareas can be different.

Fig. 7 shows a schematic view of a scanning mode of another embodiment of the invention, as a variant, with only one activatable area which can be moved within the field of view of the lidar, as indicated by the arrow, either laterally, longitudinally or in any direction. Thus, a plurality of scanning modes are preset according to one or more of the number, the size, the moving direction and the moving step length of the activatable areas, and the method can be freely configured in engineering, and is different from the previous embodiment in that: the matching sub is changed into a matching scanning mode, so that the control time sequence is simplified, and the efficiency is higher in engineering application.

According to a preferred embodiment of the invention, the matching scan pattern is activated in accordance with a change in the selected region of interest.

With continued reference to fig. 7, for example, index numbers of scan patterns may be defined, numbers of all possible options for the activatable area move. For example, there may be nine scan modes, each of which covers a different field of view, designated index 0-8, respectively. And then the user can determine the object of interest or select the region of interest according to the need by negotiating the parameters in each mode with the user, and then specify a certain index i (i epsilon 0-8) to control the movement of the activatable region.

In some embodiments, the lidar may receive random switching among 9 index, for example, selecting index 0 in the first round of scanning, acquiring the point cloud, and then switching to index 8 in the second round of scanning according to the selected region of interest of the point cloud, where the selected region of interest is located near (partially overlapping) the field of view covered by index 8, that is, if the mode of index 8 is adopted, the coverage of the region of interest may be maximized. The switching performed in this embodiment is a switching of the movement step, and the first and second scans each refer to a detection of the first field of view.

The above embodiments describe the area division and scanning mode in the laser radar field of view, and determine the distribution of the point cloud, or determine the echo data amounts of different areas, so as to dynamically allocate more or fewer processing units 13 to the detection units 12 corresponding to different areas. The advantage of the region division or scan mode is that: the dynamic performance and the cost are combined, and the engineering is difficult to achieve complete dynamic performance.

The cross-interconnection of the detection unit 12 and the processing unit 13 is hardware-based in such a region division or scanning mode that the dynamically changing echo data volume is handled with as little measurement or processing resources as possible. As described further below.

The echo and point cloud data volume of the first region is greater than the echo and point cloud data volume of the second region, and in one particular embodiment the echo and point cloud data volume of the first region occupies about 3/4 of the total echo and point cloud data volume. If the plurality of detecting units 12 and the plurality of processing units 13 are further connected by using the conventional tree MUX architecture, it is found that the processing units corresponding to the first area are not capable of processing the echo data, and the processing units in the second area are idle, that is, the problem of uneven allocation of echo processing resources occurs. Therefore, in order to reasonably allocate echo processing resources in real time, a concept of resource pool is introduced, namely, all echo processing resources are planned into one total resource pool, namely, the connection lines between the processing units 13 and the detection units 12 can be crossed, namely, two or more processing units 13 can be butted with the same detection unit 12, and the processing units 13 can share the load amount with each other according to the size of each area, the echo and the point cloud data amount.

According to a preferred embodiment of the invention, each detection unit 12 corresponding to the first zone is connected to a plurality of processing units 13.

According to a preferred embodiment of the invention, each detection unit 12 corresponding to the second area is connected to a processing unit 13. The data volume of the second area is small relative to the data volume of the first area, dynamic regulation does not exist, and the situation that the processing unit 13 connected with the second area cannot process echo data is avoided, so that each detection unit 121 in the second area is connected with one processing unit 13.

According to a preferred embodiment of the invention, each processing unit is connected to a plurality of detection units.

For the three preferred embodiments described above, further description is provided below with reference to the accompanying drawings.

Fig. 8 shows a schematic diagram of a first area walk of an embodiment of the present invention, the radar sharing x-rays, n-rays being included in the walk-in range of the first area, wherein the first area includes m-rays, and the second area includes (x-n) rays (x-n/2-rays are respectively present in the upper and lower areas of the walk-in range of the first area). The x-rays are divided into a plurality of groups of BANKs, each BANK including a plurality of detection channels, and are subdivided into odd and even BANKs according to the number of BANKs, the difference in vertical angle between every 2 lines being a vertical angle resolution, taking even BANKs as an example, the even BANKs including x/2 lines, the even BANKs including n/2 lines in the first region wandering range, and the upper and lower regions thereof being (x-n)/4 lines, respectively.

The horizontal and vertical resolutions of the first region are y times that of the other regions except the first region, and the total is y ² And (5) multiplying the point cloud data. The first region is capable of wandering within the n-line x vertical angular resolution (e.g., 0.1 °) of the vertical field of view, the current first region being a-line x 0.1 °/line in width, the step of movement being b-line/time, walking from lowermost to uppermost, the number of steps required= (n-line-a-line)/b=c steps, thus wandering throughout the dashed region, together corresponding to (c+1) modes (the current mode being the first region being lowermost in the dashed region).

Fig. 9a-9c show schematic diagrams of the connection of the BANK and the processing unit corresponding to fig. 8 according to various embodiments of the present invention, in which, in general, the number x of lines of the radar is determined, and what is needed to be optimized is the number of processing units 13 and the connection of the processing units 13 and the detection units 12. Specifically, the x-rays are divided into a plurality of groups of BANKs, each of which includes a plurality of probe channels, and can be subdivided into odd BANKs and even BANKs according to the number of BANKs.

In fig. 9a, taking even-numbered BANKs as an example (odd-numbered BANK symmetrical design), 8 BANKs are respectively BANKs 0-14, wherein BANK 0 and BANK 14 correspond to a second area, BANK 2-12 correspond to a wander range of a first area, and it is determined how echo data of detectors of which BANKs are allocated to a processing unit 13 connected thereto for processing according to a scan pattern. The number of the detecting units 12 and the transmitting units 11 in each BANK is not limited in this embodiment, for example, as an exemplary embodiment, BANK 2 includes 8 detecting units 12 and 8 transmitting units 11 corresponding thereto, BANK 8 includes 12 detecting units 12 and 12 transmitting units 11 corresponding thereto, and transmitting units 11 in the same BANK operate in a time-sharing manner, for example, one transmitting unit 11 is selected from each BANK at a time, and emits light in parallel with transmitting units 11 from other BANKs.

According to a preferred embodiment of the present invention, as shown in fig. 3c, the lidar 10 further comprises a plurality of multiplexing units 14, the plurality of detection channels are divided into a plurality of groups of BANKs, the detection units 12 of each group of BANKs are connected to one processing unit 13 through one multiplexing unit 14, and the detection units 12 of the plurality of groups of BANKs are connected to one processing unit 13 through one multiplexing unit 14.

The technical idea of the present invention is to form a cross-interconnect network between the BANK and the processing unit 13. The cross-connect is different from the fixed connection through the tree MUX shown in fig. 2, and means that each BANK may be connected to a plurality of processing units 13, and any one processing unit 13 may also be connected to a plurality of BANKs. In other words, the plurality of processing units 13 and the plurality of BANKs are connected by a cross MUX network, so-called cross indicates that at least one processing unit 13 is connected to the plurality of BANKs, and at least one BANK is connected to the plurality of processing units 13. In this way, the connection relationship of the BANK and the processing unit 13 is relatively dynamic and not fixed. Reference may be made to fig. 9a-9c.

In addition, each BANK may include a plurality of probing channels, and tree-shaped muxes may be selected among the plurality of probing channels to make multiple choices, that is, the plurality of probing channels are routed to each specific probing channel through a MUX wheel, and the plurality of probing channels work in a time-sharing manner and transmit and receive in turn. However, the invention does not limit the number of BANK and the number of detection channels in each BANK, and does not limit the basis and mode of area division.

According to a preferred embodiment of the present invention, the minimum number of processing units 13 to be connected to the detection unit 12 satisfying each scanning mode is determined according to the expected point cloud result corresponding to the preset plurality of scanning modes.

With continued reference to fig. 9a, the lidar includes four processing units, processing units 13A-13D. In a certain detection, the sequence of processing resource allocation may be matched with the scanning mode of the first area, for example, BANK 6 is connected with all of 13A, 13B and 13C, and it is possible that in a certain detection, the total point amount generated by BANK 6 is 3, and in a relatively smaller detection, echo data of BANK 6 is all allocated to 13A for processing; in the next detection, if the total point amount generated by BANK 6 is 20 and relatively large, a part of echo data of BANK 6 is allocated to 13B for example 8 and another part is allocated to 13C for example 12. It should be noted that, a detection process, that is, one transmission and one reception, may take a certain period of time, and during this period of time, the amount of echo data that each processor can process is limited, but the amount of echo data that each processor can process may be different. The total echo data which can be processed by all processors in the period is not less than the total echo data which can be responded and output by all detectors in the period.

Fig. 9a shows a connection relationship of four processing units 13 sharing echo data of even-numbered BANK, and according to the technical concept of the present invention, the connection may also be performed according to the embodiments of fig. 9b and 9c, which is different from the embodiment of fig. 9a in that: first, fig. 9b shares echo data of even BANK with three processing units, fig. 9c shares echo data of even BANK with five processing units, and if only pursuing to deal with changes in point cloud data rate with as few processing units as possible, fig. 9b is optimal with respect to fig. 9a and 9 c; second, fig. 9C shows that the middle 13C and 13D are paired from the view of the processing units, and the two processing units 13 are connected to the same BANK 6, BANK 8 and BANK 10, which means that the 13C and 13D assist each other to share echo data from the three BANKs. The upper and lower edges 13A and 13E are self-care, not paired with other processing units 13, and the butted BANK is partially located in the second area, so that the data volume is relatively small, and the echo data processing can be independently completed. Thus, for a BANK with a relatively large echo data volume during the scan period (which may include one or more transmissions and corresponding receptions, which may take a certain period of time), the paired processing units 13 are provided to achieve a mutual assistance effect, and if this is only concerned, fig. 9c is preferred over the solutions of fig. 9a and 9 b. In practice, however, the number of processing units 13 and the connection between the BANK and the processing units 13 also take into account other limiting factors, as will be further described below.

According to a preferred embodiment of the invention, the processing unit 13 connected to each detection unit 12 is determined on the basis of the echo data processing time length of each detection unit 12.

The number of processing units 13 and the connection of the BANK to the processing units 13 are dependent on the maximum load of each processing unit 13 or the number of echoes of the detection channel that can be processed. If the processing time of the processing unit 13 is divided in time slices, the time slices correspond to the echo processing duration given to each probe channel, and specifically include the probe channel switching time, the encoding time, and the flight time. In long range radars, the time of flight is the largest. For example, in a radar with a range of 300m, it is necessary to allocate hundreds of ns (channel switching time) +hundreds of ns (coding time) +2000ns (flight time=300 m×2m/3×10) ⁸ m/s) =thousands of ns of time slices to each detection channel. Thus, the length of time for echo data processing needs to be taken into account when determining the connection of the probe unit 12 to the processing unit 13.

According to a preferred embodiment of the invention, the rotational speed of the lidar 10 and the point cloud resolution also have to be taken into account when determining the processing unit 13 connected to each detection unit 12.

The number of processing units 13 and the link relationship of the BANK to the processing units 13 are also limited by the rotational speed of the radar and the point cloud resolution. Once the rotation speed and resolution of the radar are fixed, for example, the horizontal resolution is 0.1 degrees, the vertical resolution is 0.2 degrees, and x detection channels are required to be detected correspondingly, the e detection channels are required to be transmitted and received within the period of 0.1 degree of rotation of the radar. The faster the rotation speed, the shorter the time period given to each detection channel, and thus the parallel detection of f (f < e) detection channels is possible, and the processing units 13 share the processing of echo data of the f channels with each other. For example, the radar scans the horizontal field of view under the drive of the rotating mirror, the scanning frequency is 5Hz, if g° (relatively coarse horizontal angular resolution) is required, then (1 s/5×360 °) ×g° =hus is required, and then hus is one scanning period. If a rotation of i (finer horizontal angular resolution, i < g) is required, j us being the period of one scan.

According to the ranging and timing plan, the time slice required for each detection channel to complete detection is d us (=the duration for which the processing of echo data of one detection channel is given), so that the maximum point cloud load of each processing unit 13 is h/d in this scanning period of h us.

If the user has a high resolution requirement for the tracking area, for example, the number Points of the point clouds to be detected for each area is determined within h us, the walk range for the first area is n lines, and the BANK for the first area corresponds to the pattern index, so that it is necessary to complete h/d point clouds within h us, i.e. to divide the sum of the numbers to be processed by all the processing units 13, for one full-load BANK in the first area. At the same time, this number is exactly equal to the maximum number of clouds that each processing unit 13 can receive in h us. Therefore, one BANK may need 1 processing unit 13 for the BANK corresponding to the first area, and the processing unit 13 cannot serve the processing of echo data of other BANKs.

With continued reference to fig. 9a-9c, it will be appreciated that the most central of the first region of travel, such as BANKs 6, 8, 10, is associated with a plurality of processing units 13, rather than one processing unit 13, because the first region is traveling and peak data (e.g., 36) is reached, the echo data of one BANK will need to occupy one processing unit 13, and of course the echo data of another BANK will need to be split into another processing unit or units 13 for processing.

From the above analysis, the case of processing resources is: the processing units 13 are not in one-to-one correspondence with the BANK, the number of processing units 13 is limited, and the wiring rules of fig. 9a-9c are: each processing unit 13 is connected to a plurality of BANKs, each BANK being connected to a plurality of processing units 13, the final aim being to ensure: in either area division or scanning mode (representing the distribution of echo and point cloud data amounts), the echo data output by the correspondingly operated detection unit 12 can be processed by a relatively limited processing unit 13.

It is theoretically possible to have each BANK connected to a processing unit and each processing unit connected to each BANK, but it is preferable that a factor is also considered that: layout arrangement.

According to a preferred embodiment of the invention, the processing unit 13 connected to each detection unit 12 is determined on the basis of the physical distance between the detection unit 12 and the processing unit 13.

The number of processing units 13 and the wiring relationship of the BANK to the processing units 13 are also limited by layout. Specifically, the more wires that need to be connected, the more complex the layout of the plurality of BANK and the plurality of processing units 13 on the circuit board, if each of the detection units 12 is connected to all of the processing units and each of the processing units 13 is connected to the detection units according to the theoretical operation, it is limited by: 1) The area of the circuit board is limited, and the circuit board is not enough for arranging all the connecting wires; 2) The wires may cross each other, causing signal crosstalk; 3) The heat dissipation is not smooth and the reliability is poor.

Therefore, the limiting factor is that each BANK or processing unit is connected with the processing unit or BANK in a range which is relatively close to the processing unit or processing unit as far as possible, and the condition that the connection is too many or the cross line is too many is avoided as far as possible.

The number of processing units and the limiting factors of the connection relation between the BANK and the processing units are described above, in the section describing the area division and the scanning modes, for example, 9 scanning modes are mentioned for the activatable areas, the coverage range of the activatable areas in each scanning mode is different, the internal control timing and the point cloud format are different correspondingly, and the activation mode between the BANK and the processing units can be different. Therefore, after the connection between the BANK and the processing units is completed by comprehensively considering the limiting factors, how to allocate the echo data output by each detector in the period to a proper amount of processing resources to complete the point cloud computing and generating can be planned on the premise of the data volume of each scanning mode and the data volume bearable by the MUX.

Fig. 10a is a schematic diagram showing a scan pattern index=1 connection and an echo data allocation manner according to an embodiment of the present invention, in which diagonal line filling is represented as a current first area, broken line represents a hardware connection relationship, solid line represents a connection to which echo data is actually allocated, and a number on each solid line represents a point cloud number completed by a corresponding processing unit in one period, for example, h=90 us. Only the even BANK designs are listed in fig. 10a, and the odd BANKs are symmetrical designs, and are not described or shown here again. Wherein the point cloud number of the first region is g=3 times as large as the other regions except the current first region in the range of the first region wander in both the horizontal field of view and the vertical field of view, so the total number is g ² =9 times. Since the point clouds of BANK 4 and BANK 6 reach the peak (e.g. 36), it is necessary to correspond to one processing unit 13 each, that is, to process the whole echo data of BANK 4 by 13B and to process the whole echo data of BANK 6 by 13C. While the point cloud of the other BANK is smaller, 13A, 13D and 13E can respectively correspond to two BANK, namely the total echo data from the detectors of 2 BANK can be processed in the period of time.

Fig. 10b is a schematic diagram showing a scan pattern index=9 connection and an echo data allocation manner according to another embodiment of the present invention, in which diagonal line filling is represented as a current first area, broken line represents a hardware connection relationship, solid line represents a connection to which echo data is actually allocated, and a number on each solid line represents a point cloud number completed by a corresponding processing unit in one cycle. Also taking an even BANK as an example, since the point clouds of BANK 10 and BANK 12 reach a peak (e.g., 36), it is necessary to correspond to one processing unit 13 each. While the point clouds of the other BANKs are smaller, 13A, 13B and 13E can share the processing, i.e. the total amount of echo data from the 3 BANK detectors of 13A, 13B and 13E can be processed during this period.

According to a preferred embodiment of the present invention, a plurality of mapping tables are preset according to the point cloud distribution, where the mapping tables include a range of a first area, and corresponding processing resources are allocated to each detection unit 12 in the first area according to the echo data amount, the total echo data amount output by all detection units in the period of time, and the upper echo limit (processing resource) that each detection unit can process in the period of time.

Different cross MUX mapping tables are designed according to the point cloud distribution, the processing time of each processing unit 13 is divided according to time slices, and the time slices are distributed according to the MUX mapping tables, so that a plurality of processing units 13 are used as a resource pool. Specifically, time slices are allocated to different BANKs, so that detection channels in different BANKs can have different point cloud data rates to support movement of the first area in a vertical direction (or any direction). The detection channel occupied by the first area may generate at least one time more point cloud data than the second area, and the detection channel is allocated to a relatively more time slice of the BANK corresponding to the first area and a relatively less time slice of the BANK corresponding to the other area.

The mapping table includes a range of the first area, and an upper echo limit (processing resource) that each detection unit can process in the period of time according to the echo data amount, the total echo data amount output by all the detection units in the period of time. For example, according to a preset scanning mode of the first area in the vertical direction, a fixed cross MUX mapping table is correspondingly set. The table comprises a plurality of time slices, each time slice is internally provided with an address of each cross MUX and a tree-shaped MUX address inside the BANK, so that a detection channel of work in each time slice can be determined, and further, the working time of each point cloud and a corresponding horizontal angle can be determined.

According to a preferred embodiment of the invention, the matching mapping table is switched according to the change of the point cloud distribution.

Different cross MUX mapping tables are preset according to the point cloud distribution, and then the mapping tables can be selected and switched to be matched according to the change of the point cloud distribution, so that the dynamic tracking of the obstacle or the target area is realized.

For example, the index number of the scanning mode is specified by the user, and the lidar automatically switches the corresponding cross MUX mapping table and the point cloud format, so that the first area can be moved in real time in the vertical direction.

According to a preferred embodiment of the invention, the mapping table is switched between two probes.

In summary, the above description has been made by way of embodiments of region division, scan mode, point cloud distribution of each region, hardware connection relation between the processing unit 13 and the detecting unit 12, and how to allocate processing resources, and the invention comprehensively considers hardware overhead and algorithm implementation, dynamically tracks the resolution and the change of the point cloud data rate caused by one or more regions with few measurement resource processing, and is particularly suitable for the requirement of balancing the cost, volume and efficiency in engineering.

According to a preferred embodiment of the present invention, referring to fig. 3a, the lidar 10 comprises a plurality of transmitting units 11, a plurality of detecting units 12 and a plurality of processing units 13, wherein the plurality of transmitting units 11 are configured to transmit the detecting light beams, and the plurality of detecting units 12 and the plurality of transmitting units 11 form a plurality of detecting channels, and each detecting channel corresponds to a vertical orientation. The invention also provides a resource allocation method of the laser radar. The resource allocation method comprises the following steps:

The plurality of detection units 12 receive echoes of the detection light beams emitted by the emission unit 11 reflected by obstacles;

distributing echo data of the detection unit 12 to one or more processing units 13 connected thereto for processing the echoes and generating a point cloud;

wherein the echo processing relationship of the detection unit and the processing unit is dynamically adjustable.

According to a preferred embodiment of the invention, at least one of the probe units 12 is connected to a plurality of processing units 13, and for the probe units 12 connected to a plurality of processing units 13, echo data thereof can be shared for processing by a plurality of processing units 13.

According to a preferred embodiment of the present invention, the resource allocation method further comprises: when the field of view of the lidar 10 includes a first region and a second region, and the echo data amount in the first region is greater than the echo data amount in the second region, a processing resource capable of completing the total amount of echo data that can be generated by the detection unit 12 in the period of time is allocated to the detection unit 12 corresponding to the first region according to the echo data amount.

According to a preferred embodiment of the present invention, wherein the first region may comprise a plurality of sub-regions, the resource allocation method further comprises: and allocating processing resources capable of completing the total amount of echo data which can be generated by the detection unit 12 in the period of time to the detection unit 12 corresponding to the first area according to the amount of echo data.

According to a preferred embodiment of the invention, the plurality of sub-areas do not overlap each other.

According to a preferred embodiment of the present invention, the resource allocation method further comprises: and allocating processing resources capable of completing the total echo data amount which can be generated by the detection unit 12 in the period of time to the detection unit 12 corresponding to the second area according to the echo data amount.

According to a preferred embodiment of the present invention, the plurality of detection channels are divided into a plurality of groups, and the detection channels of each group emit light in a time-sharing manner, and the resource allocation method further includes: based on the amount of echo data, the group of probe units 12 is allocated processing resources that are capable of completing the total amount of echo data that the probe units 12 are capable of producing during the period of time.

A plurality of mapping tables are preset according to the point cloud distribution, wherein the mapping tables comprise the range of a first area, and corresponding processing resources are allocated to each detection unit 12 of the first area according to the echo data quantity, the total echo data quantity output by all detection units in the period of time and the upper echo limit (processing resources) which can be processed by each detection unit in the period of time. According to a preferred embodiment of the present invention, the resource allocation method further comprises: and switching the matched mapping table according to the change of the point cloud distribution.

According to a preferred embodiment of the present invention, the resource allocation method further comprises: the mapping table is switched between the two probes.

The invention also relates to a computer readable storage medium comprising computer executable instructions stored thereon, which when executed by a processor implement a resource allocation method as described above.

Finally, it should be noted that: the foregoing description is only a preferred embodiment of the present invention and is not intended to limit the present invention, but although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the technical solutions described in the foregoing embodiments, or equivalents may be substituted for some of the technical features thereof. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A lidar, comprising:

a plurality of emission units configured to emit probe light beams;

a plurality of detection units configured to receive echoes of the detection light beams emitted by the emission units reflected by the obstacle; and

2. The lidar of claim 1, wherein at least one of the detection units is coupled to a plurality of processing units, and wherein for a detection unit coupled to the plurality of processing units, echoes thereof are shared for processing by the plurality of processing units.

3. The lidar of claim 1, wherein when the field of view of the lidar comprises a first region and a second region, and the amount of echo data in the first region is greater than the amount of echo data in the second region, at least one detection unit corresponding to the first region is coupled to a plurality of processing units.

4. A lidar according to claim 3, wherein the first region may comprise a plurality of sub-regions, with at least one detection unit corresponding to each sub-region being connected to a plurality of processing units.

5. The lidar of claim 4, wherein the plurality of sub-regions do not overlap one another.

6. A lidar according to claim 3, wherein each detection unit corresponding to the first region is connected to a plurality of processing units.

7. A lidar according to claim 3, wherein each detection unit corresponding to the second region is connected to a processing unit.

8. The lidar of claim 1, wherein each processing unit is coupled to a plurality of detection units.

9. The lidar according to any of claims 1 to 8, wherein the lidar further comprises a plurality of multiplexing units, the plurality of detection channels being divided into a plurality of groups, the detection units of each group being connected to one processing unit by one multiplexing unit, the detection units of the plurality of groups being connected to one processing unit by one multiplexing unit.

10. The lidar according to any of claims 1 to 8, wherein a minimum number of processing units to which the detection units satisfying each scanning mode are to be connected is determined corresponding to a preset number of scanning modes based on an expected point cloud result.

11. The lidar of claim 10, wherein the processing unit coupled to each detection unit is determined based on a length of time the echo data is processed for each detection unit.

12. The lidar of claim 10, wherein the processing unit coupled to each detection unit is determined based on a rotational speed of the lidar and a point cloud resolution.

13. The lidar of claim 10, wherein the processing unit connected to each detection unit is determined based on a physical distance between the detection unit and the processing unit.

14. A resource allocation method of a laser radar including a plurality of transmitting units configured to transmit probe beams, a plurality of detecting units, and a plurality of processing units, the resource allocation method comprising:

the plurality of detection units receive echoes of the detection light beams sent by the sending units and reflected by the obstacles;

15. The resource allocation method of claim 14, wherein at least one of the probe units is coupled to a plurality of processing units, and for probe units coupled to the plurality of processing units, echoes thereof are shared for processing by the plurality of processing units.

16. The resource allocation method of claim 14, further comprising: when the field of view of the laser radar comprises a first area and a second area, and the echo data volume in the first area is larger than that in the second area, distributing the echo in the first area to one or more processing units connected with the detection units corresponding to the first area according to the echo data volume.

17. The resource allocation method of claim 16, wherein the first region may include a plurality of sub-regions, the resource allocation method further comprising: and distributing the echo data in the plurality of subareas to one or more processing units connected with the detection units corresponding to the plurality of subareas according to the echo data quantity in the plurality of subareas.

18. The resource allocation method according to claim 17, wherein the plurality of sub-regions do not overlap each other.

19. The resource allocation method of claim 16, further comprising: and distributing the echo data in the second area to a processing unit connected with the detection unit corresponding to the second area according to the echo data quantity in the second area.

20. The resource allocation method according to claim 14, wherein the plurality of probe channels are divided into a plurality of groups, the resource allocation method further comprising: the echo data of each set of detection channels is distributed to one or more processing units connected to the set of detection channels according to the echo data volume of the detection channels.

21. The resource allocation method of claim 16, further comprising: and presetting a plurality of mapping tables according to the point cloud distribution, wherein the mapping tables comprise a range of a first area and processing units connected with detection units corresponding to the first area, and distributing echo data in the first area to the processing units connected with the detection units corresponding to the first area according to the echo data quantity in the first area.

22. The resource allocation method of claim 21, further comprising: and switching the matched mapping table according to the change of the point cloud distribution.

23. The resource allocation method of claim 22, further comprising: the mapping table is switched between the two probes.

24. A computer readable storage medium comprising computer executable instructions stored thereon, which when executed by a processor implement the resource allocation method of any of claims 14-23.