JP2021514457A - Rain filtering techniques for autonomous vehicles - Google Patents

Abstract

The present invention relates to systems and methods for using a rider to navigate an autonomous vehicle during rain or snow conditions. Two rider units are used to scan the area close to the autonomous vehicle at the same time. The laser light reflected from the two rider devices is analyzed using an algorithm. The reflected laser light, undetected by both units, can be due to rain. The rider unit can use multi-echo scanning. In addition, the rider device can scan a series of planes (ie, multiplanar scan) to improve the robustness of the system. The system improves the reliability of rider navigation when rain, snow or other particles are present in the air.

Description

The present invention relates to systems and methods of using a rider for autonomous vehicle navigation when particulate matter, such as rain or snow, is present in the air.

Autonomous vehicles can monitor their environment and navigate without human input. The benefits of autonomous vehicles include greater vehicle safety, greater efficiency, lower costs, reduced congestion, and greater mobility for unsuitable or unlicensed people.

Autonomous vehicles use sensors to monitor their location and their environment. CMOS cameras, radar and laser light can be used to detect objects and dangers around them. However, each type of sensor has its limitations. For example, CMOS cameras rely on sufficient lighting to detect objects. Moreover, the camera may not be effective, for example, when light is projected backwards by the headlights of an approaching vehicle. Other drawbacks of the camera include inadequate lighting, blurring, limited field of view and / or detection range. Radar does not rely on visible light, but does not have the resolution required to identify smaller objects. Laser light can be used to overcome these limitations, but may not be effective during rain or snow.

Lidar (Light Imaging Detection and Ranger) measures the distance to an object by illuminating the object with a pulsed laser beam and measuring the reflected pulse with a sensor. To do. In general use, the rider system reflects multiple laser pulses from objects surrounding the vehicle. Laser return time differences and wavelengths can then be used to create a digital 3D representation of the object.

However, it is difficult for the rider to penetrate snow, rain, fog and dust in the air. Rain or other particles in the air can reflect the laser light back into the system, similar to a solid object. As the pulsed laser light is reflected back into the rider system, the autonomous vehicle unnecessarily slows down or stops. This is due to the "false positive" detection of obstacles.

Methods have been proposed to improve the rider's robustness in rainy or snowy conditions. For example, a rider sensor using "multi-echo" or "last echo" technology can filter the reflections caused by dust, rain and fog. They operate on the premise that some of the energy from the pulse can be reflected by the particulate matter. The remaining beam can continue to propagate and is reflected by solids. When this happens, the rider unit can ignore the closer and weaker reflections caused by rain or fine particles in the air. While improving, this technique is not robust for filtering moderate to heavy rain.

A method of combining a radar with a rider to detect an object is described (see Patent Document 1). The rider is used to create a 3D point map in the area surrounding the vehicle. Radar is then used to confirm that the object or danger is a solid material rather than rain or aerial particles. The system is designed to prevent false positive rider detection, but it requires the use of a separate radar system. In addition, radar may miss objects with inherently low resolution and weak reflectance.

A method of using a rider to confirm the existence of an object close to an autonomous vehicle is described (see Patent Document 2). The rider detects multiple points on the object (ie right, left, up and down) to confirm the presence of the object. The method is similar to the multi-echo method because it requires multiple pulses. Similar to Patent Document 1, the multi-echo method may not be effective in detecting small objects.

U.S. Pat. No. 9097800 U.S. Patent Application No. 2014/576265

Therefore, there is a need for a system that can allow autonomous vehicles to navigate using the rider in the rain or snow.

The present invention recognizes the need for improved systems for autonomous vehicles to navigate and detect objects and hazards. The system should be reliable and able to operate in all weather conditions. The present invention includes a system and a rain filtering algorithm to allow an autonomous vehicle to operate in rain or snow.

The following summary is provided to facilitate understanding of some of the groundbreaking features inherent in the disclosed embodiments and is not a complete explanation. A good understanding of the various aspects of the embodiments disclosed herein can be obtained by taking into account the entire specification, claims, drawings, and abstract as a whole.

The present invention includes systems for detecting and / or navigating objects using laser light (ie, lidar), and for distinguishing solids from particles in the air. The system emits a pulse of laser light and detects the reflected laser light with a first rider device that scans the area and emits a pulse of laser light to detect the reflected laser light. This can include a second rider device that scans the same or substantially the same area. The computer uses the laser light emitted by the first rider device and the detected reflected laser light to confirm the presence of the solid matter, and the laser light emitted by the second rider device and detection. An algorithm can be used to compare with the reflected reflected laser light. The presence of a solid object can be confirmed when the first and second rider devices detect the laser light reflected from the solid object. The first and second rider devices can be mounted on autonomous vehicles such that they are 0.5 to 2.5 meters apart from each other on a horizontal plane.

In addition to the dual rider unit technology, the first rider device and the second rider device can utilize multi-echo technology to distinguish solids from aerial particles. In addition, the lidar device can perform multiple plane scans to distinguish solids from aerial particles.

The present invention also includes methods for detecting objects or obstacles for autonomous driving. The first rider device scans an area close to an autonomous vehicle by emitting laser light and detecting the reflected laser light (ie, echo). The second rider device scans an area close to the autonomous vehicle by emitting laser light and detecting echoes. Using computers and algorithms to detect and confirm the presence of objects or obstacles, the pulses emitted by the first rider device and the detected echoes are emitted by the second rider device. It is compared with the pulse and the detected echo. The presence of a solid object can be confirmed when both the first rider unit and the second rider unit detect echoes from the object.

Echoes caused by rain or snow can be identified by comparing the echoes received by the first rider device with the echoes received by the second rider device. The first rider device and the second rider device can use multi-echo scanning. In addition, the rider device can scan a series of layers or planes. The presence of a solid can be confirmed when it appears in one or more layers or planes.

The present invention is also a method of detecting a solid object, wherein (a) a first lidar unit is used to emit a pulse of laser light, and an echo is detected to scan an area close to an autonomous vehicle. Steps and (b) traversing substantially the same area by emitting a pulse of laser light using a second rider unit and detecting echoes, and (c) the presence of an object or obstacle. An algorithm for comparing the pulses emitted and received echoes by the first rider unit with the pulses emitted and received echoes by the second rider unit to detect and confirm Includes a method consisting of steps to use. False positives (ie, particulate matter in the air, such as rain or snow) can be identified when one of the rider units detects an echo but the other rider unit does not detect an echo.

In addition, the first rider unit and the second rider unit can use multi-echo scanning. They can also scan a series of planes. The presence of a solid object can be confirmed when the object appears in more than one plane. In addition, the presence of solids can be confirmed in the last echo of the multi-echo scan.

The present invention is also a method for detecting and confirming the presence of a solid object using a lidar, a) a plurality of planes in which a pulse of laser light is emitted and the reflected laser light is detected. It relates to the step of using the first rider device to scan the b) the laser light emitted from the first rider device and the reflected laser light detected by the first rider device. By comparing multiple planes for the steps of collecting data and c) the presence and / or absence of reflected light, the light reflected from the first rider device is the result of particulate matter in the air. The step of determining if there is, d) the step of using the second rider device to scan multiple planes where the pulse of the laser light is emitted and the reflected laser light is detected, and e. ) The step of collecting data related to the laser light emitted from the second rider device and the reflected laser light detected by the second rider device, and f) the presence of the reflected laser light and / Or the step of determining whether the light reflected from the second rider device is the result of particulate matter in the air by comparing multiple planes for absence, and g) the reflected light is air. With the steps of using the algorithm to compare the data from the first rider device with the data from the second rider device to ensure that it is the result of a solid rather than the particulate matter inside. Includes methods consisting of. The planes scanned by the first rider device are in the same area or substantially the same area as the planes scanned by the second rider device. Solid when reflected laser light is detected within four or more of the planes of the first rider device and / or within four or more of the planes of the second rider device. The existence of an object can be confirmed. In addition, the first rider device and the second rider device can use multi-echo scanning.

Introduction A first aspect of the invention is an improved system and method for using a rider to detect obstacles for autonomous driving.

A second aspect of the invention is an improved system and method for autonomous driving utilizing multiple rider systems.

A third aspect of the invention is an improvement for autonomous driving utilizing multiple rider systems and algorithms to avoid erroneously identifying water or particles in the air as solids. System and method.

A fourth aspect of the invention is an improvement for autonomous driving that utilizes multi-echo technology to scan multiple layers and employ algorithms to mitigate interference caused by rain or particles in the air. System and method.

A fifth aspect of the invention utilizes dual lidar systems, multi-echo technology and multi-layer scanning with algorithms to improve robustness and mitigate interference caused by rain or particles in the air. , Improved systems and methods for autonomous driving.

The drawings described herein are intended to illustrate only selected embodiments, not all possible implementations, and are not intended to limit the scope of the present disclosure. ..

The above overview, as well as the following detailed description of the exemplary embodiments, will be better understood when read in conjunction with the accompanying drawings. Illustrative configurations of the present disclosure are shown in the drawings for purposes of explaining the present disclosure. However, the disclosure is not limited to the particular methods and instruments disclosed herein. In addition, the drawings are not at a constant scale.

FIG. 1A is a diagram showing a conventional multi-echo rider system. FIG. 1B is a diagram showing a multi-layer scanning lidar in which each scanning layer is superimposed on a grid map. FIG. 2 is a diagram showing a grid obtained by a multilayer scanning lidar. Solids exist on multiple layers. FIG. 3A is a flowchart of an embodiment in which the rider sensor uses last echo filtering and a rain filter algorithm for obstacle feature detection. FIG. 3B is a diagram showing a dual rider unit used to detect an obstacle according to an embodiment of the present invention. Figure 3C shows a book in which the dual rider sensor uses last echo filtering and a rain filter algorithm to distinguish the return signal caused by rain or particles in the air from the return signal of a solid. It is a flowchart of one Embodiment of the invention. FIG. 4A is a diagram showing an autonomous vehicle having dual rider sensors separated by a distance of "d". FIG. 4B is a diagram showing dual rider sensors and their respective zones of detection. FIG. 5A is a plot of a single plane rider. FIG. 5B is a plot of a single plane rider using a dual rider system. FIG. 5C is a plot of a multi-plane rider. FIG. 5D is a plot of a multi-plane rider using a dual rider system.

Definitions Although the present invention is described primarily with respect to autonomous vehicle navigation, it should be understood that the present invention is not so limited and may be used to support other efforts to use the rider. Other applications include using the invention in other vehicles, such as robots, drones or unmanned aerial vehicle systems. The present invention can also be used to improve the robustness of the rider in landscape imaging and mapping applications when particles such as rain or snow are present in the air.

References herein to "one embodiment / aspect" or "execution / aspect" are at least one embodiment / aspect of the present disclosure in which the particular feature, structure, or property described with respect to the embodiment / aspect is described. Means to be included in. The use of the phrase "in one embodiment / embodiment" or "in another embodiment / embodiment" in various places throughout the specification does not necessarily mean all of them refer to the same embodiment / embodiment. Also, separate or alternative embodiments / embodiments do not exclude other embodiments / embodiments from each other. Moreover, various features that are shown in some embodiments / embodiments and may not be shown in other embodiments / embodiments are described. Similarly, various requirements are described that are requirements in some embodiments / embodiments but may not be requirements in other embodiments / embodiments. The embodiments and embodiments may be used interchangeably in some cases.

The terms used herein generally have their usual meaning in the art, within the context of the present disclosure, and in a particular context, where each term is used. Some terms used to describe this disclosure will be described below or elsewhere herein to provide additional guidance to those skilled in the art with respect to the description of this disclosure. For convenience, some terms may be highlighted using, for example, italics and / or quotation marks. The use of highlights does not affect the scope and meaning of the term. That is, the scope and meaning of a term is the same in the same context, whether it is highlighted or not. It will be understood that the same can be said in more than one way.

Therefore, alternative wording and synonyms may be used for any one or more of the terms described herein. Nor should there be any particular importance as to whether the term is detailed or explained herein. Synonyms for some terms are given. The description of one or more synonyms does not preclude the use of other synonyms. The use of examples anywhere in the specification, including examples of any of the terms described herein, is merely exemplary and is either disclosed or exemplified herein. Does not limit the scope and meaning of the terms. Similarly, the present disclosure is not limited to the various embodiments given herein.

Without further limiting the scope of this disclosure, examples of instruments, devices, methods and related results thereof according to embodiments of the present disclosure are given below. Please note that titles or subtitles may be used in the examples for the convenience of the reader and should never limit the scope of this disclosure. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this disclosure relates. In case of inconsistency, this document, including the definition, controls.

Although not limited, directional terms and / or related terms such as left, right, nadir, apex, top, bottom, vertical, horizontal, back, front, and lateral are relative to each other and applicable elements. Alternatively, it depends on the particular orientation of the article and is therefore used to aid in the description of the various embodiments and the appended claims herein and is not necessarily construed as limiting. Not exclusively.

Where applicable, the terms "about" or "generally" mean a +/- 20% margin when used herein and in the appended claims, unless otherwise specified. .. Also, where applicable, the term "substantially" means a margin of +/- 10% when used herein and in the appended claims, unless otherwise specified. .. It should be understood that all uses of the above terms are not always quantifiable so that the referenced scope can be applied.

The term "multi-echo function" refers to the ability of a rider unit to collect and evaluate multiple (eg, 3) echoes per transmitted laser pulse. When the echo reaches the receiver of the unit, the received intensity is converted to voltage. Echoes from solids usually result in high voltages over a long period of time. Raindrop echoes, however, result in extremely low voltages over a short period of time.

The term "multilayer technology" refers to a rider unit that allows pitch angle compensation with multiple (eg, four) scanning surfaces with different vertical angles. In a preferred design, the photodiode receiver of the unit includes a plurality of (eg, four) independent receivers arranged in a row. Each receiver scans a single plane, thereby dividing the vertical aperture into multiple planes.

The term "SLAM" refers to Simultaneous Localization And Mapping, which enables accurate mapping when GPS localization is not available, such as in an indoor space. The SLAM algorithm simultaneously uses LiDAR data and IMU (Inertial Measurement Unit) data to locate the sensor and generate a coherent map around the sensor.

The term "Time-of-Flight Principle" is used between a sensor and an object based on the time difference between the emission of the signal and the return of the signal to the sensor after it has been reflected by the object. Refers to a method for measuring the distance of.

Other technical terms used herein have their usual meaning in the art in which they are used, as exemplified by various technical term dictionaries. The particular values and configurations described in these non-limiting examples may vary and may be cited solely to indicate at least one embodiment, without limiting the scope of that embodiment.

Description of Preferred Embodiment The rider is basically a distance measuring technique. The rider system sends light energy to the ground or towards an object. This emitted light can be called a "beam" or "pulse". The rider unit measures the light that is reflected back into the sensor. This reflected light can be called an "echo" or "return". The spatial distance between the rider system and the measurement point is calculated by comparing the delay between the pulse and the return.

Riders are often the preferred system for use in autonomous vehicles because they can accurately map the three-dimensional perimeter of the vehicle to high resolution. However, particulate matter in the rain or air can make the rider ineffective, as the particles reflect the laser light back into the rider unit. When the light is reflected, the autonomous vehicle is delayed or stopped because the airborne particles are indistinguishable from solids.

FIG. 1A shows a rider unit 110 that uses last echo filtering to detect an object 120. This technique is also called multi-echo scanning. Rain 115 is shown between the rider unit 110 and the object 120. One pulse can generate multiple echoes of laser light (ie, first echo, intermediate echo, last echo) for the rider unit. This can be done when particles in the air, such as rain, reflect the laser light and return it to the rider unit 110. A first echo 130 is shown in which the signal is reflected from particles in the air. Intermediate echoes can also occur as a result of particles in the air. The last echo 140 is shown, where the signal (ie echo) is reflected from the solid 120.

The system can analyze echoes to identify echoes that are the result of particles in the rain or air. With the "last echo filter", the first and intermediate echoes are due to particles in the rain or air. The last echo is due to a solid object. However, this technique has limitations and is ineffective in heavy rain or in the presence of substantial particulate matter in the air.

FIG. 1B shows a multi-layer scan, an alternative technique aimed at identifying echoes or returns from rain 115 or particulate matter in the air. The rider unit 110 scans a plurality of corners. Layers, planes or stacked planes 130 can result from scanning at each angle. The system can compare data from layers to confirm the presence or absence of objects. Objects appearing on a single layer can be attributed to particles in the air. In a preferred method, the presence of solids is confirmed when echoes are detected in four or more adjacent layers.

Each scan angle / layer can be overlaid on a grid map as shown in FIG. The results of the scan are represented on four layers. Of the four scanned corners, each corner is shown on a horizontal grid. Each shaded square indicates that the rider unit has detected an echo. Conventional rider systems show the presence of two objects 215 and 225. Analysis of data from multiple layers can distinguish false positive reads from true reads.

If obstacles are detected in scanning layer 1 but not in exact location in scanning layer 2 or N, those obstacles can be identified as particles in the air, such as rain. The first object 215 exists in the plurality of layers, but does not exist in the intermediate layer (scanning layer 3). In addition, the location of the first object 215 varies between layers. In this situation, the system can conclude that the echo is due to rain or particles in the air. In contrast, the second object 225 resides in the same location in multiple continuous layers. Using this data, the system can confirm that the signal is due to solids. In a preferred method, echoes in four cells are used to confirm the presence of an object or obstacle. If the object is in three or fewer cells, the echo is due to rain or airborne particles.

FIG. 3A is a flow chart showing a rider unit that employs both last echo filtering and rain filter algorithms. The rider unit can use a last echo filter 305 and a rain filter algorithm 315 that analyzes data from multi-layer scans for obstacle features and detection 325. This improves rider reliability and robustness in the presence of aerial particles such as rain.

FIG. 3B shows the invention in which two rider units function simultaneously to improve the robustness of the autonomous vehicle navigation system, allowing the autonomous vehicle navigation system to operate in rainy or snowy conditions. An embodiment is shown. The system can include a first (ie, left side) rider unit 210 and a second (ie, right side) rider unit 220 that scans substantially the same area. Data can be collected from each rider unit and compared using algorithms to ensure that the signal is the result of a solid object.

When the first rider unit receives the light reflected from the object, the system uses the second rider unit to confirm the presence of the object. The system can use algorithms to analyze data from each rider unit, including emitted light, reflected light (ie echo), and angle of reflection and movement of objects. In this way, the second rider unit confirms that the reflected light received by the first rider unit is the result of a solid object. The reflected pulse (ie echo) detected by one of the units can be classified as false positive due to particles in the air without confirmation from the other. The following principles can be applied to this method. That is,
-Solid objects / obstacles are detected by both rider units (210, 220).
Echoes detected by the left rider unit 210 but not by the right rider unit 220 can be due to particles in the air, such as rain.
The echo detected by the right rider unit 220 and not by the left rider unit 210 can be due to particles in the air such as rain.

This dual unit technique can be combined with other filter techniques to improve its robustness and reliability, such as last echo filters and / or multi-layer scanning. FIG. 3B shows the first echo 130 along with the last echo 140.

FIG. 3C is a flowchart showing a system of two rider units according to an embodiment of the present invention. The first rider unit 210 and the second rider unit 220 can use last echo filtering (310, 320). The data from each unit is collected, processed and analyzed using the rain filter algorithm 330. The algorithm can attribute the echo to rain, for example, if the echo is detected by a single rider unit rather than a pair. This leads to a more robust obstacle and feature detection 340. Autonomous vehicles can avoid objects / hazards without being delayed and / or stopped by false positive signals from particles in the rain or air.

FIG. 4A shows a preferred configuration of a system in which a first rider unit 210 and a second rider unit 220 are mounted on the roof of an automobile 410 and separated from each other by a distance of "d". The distance "d" can be determined empirically and is preferably between 0.5 and 2.5 meters. However, the rider units can be separated by variables depending on variables such as vehicle size, rider unit resolution, expected object / danger distance, and amount of particulate matter in the air. The ideal distance can vary, for example, from 1 cm for small mobile robots to several meters for large drones. As can be understood, the first rider unit 210 and the second rider unit 220 can be mounted at other external locations on the vehicle (eg, hood bumpers). In a preferred configuration, the rider units are placed in a horizontal plane with each other so that the laser beams are emitted in the same plane.

FIG. 4B shows the detection zone 400 of the first (ie left side) rider unit 210 and the second (ie right side) rider unit 220. Each system scans a substantially circular area to detect the object 120. Scans overlap in central area 235. The number and direction of light rays emitted by each sensor can vary based on the rider unit and settings.

The unit scans substantially the same area, but the data collected from the overlapping areas can be analyzed using algorithms to ensure that the signal is the result of solids. Particles in the rain or air can reflect laser light to the rider unit. The solid 120 is detected from the laser light reflected by both rider units.

The two rider units work at the same time and the echo is analyzed. Multi-unit technology can be combined with last echo filtering techniques and multi-layer scanning to improve the robustness of autonomous vehicle navigation and allow it to operate while particles such as rain or snow are present in the air.

Example Rider use for vehicle navigation in the rain This system was tested using an autonomous vehicle in a rain tunnel with a steady flow of rain 10 mm (mm) per hour. The rider units were fixed to the roof of the vehicle at a distance of about 1 meter from each other. Two types of riders were used: a single plane (LMS 151 unit) and a four plane (LDMRS unit). The rider unit used multi-echo technology.

5A-5D are top view representations of an environment with an autonomous vehicle in the center. The horizontal axis represents the distance in meters away from the front and rear surfaces of the vehicle. The vertical axis represents the distance in meters away from the side of the vehicle. Each segment represents a distance of 5 meters.

FIG. 5A is a plot of a single plane lidar using multi-echo scanning. The area of the circle is the known location of solid objects (pillars in the rain tunnel). Solids are detected with additional areas from false positives from raindrops 505. Most of the false positives are detected in the area in front of the vehicle.

The test was repeated using a (single plane) dual rider system in which the first rider system and the second rider system scan the area. FIG. 5B is a plot of a single plane rider with data from a dual rider system. Only scan points detected by both the first rider system and the second rider system are included. Similar to FIG. 5A, the horizontal axis represents the distance from the front and rear surfaces of the vehicle. The vertical axis represents the distance from the side of the vehicle.

The return signal due to rain can be identified by comparing FIGS. 5A and 5B. Without the dual rider system, false positives exist and are due to echoes from raindrops 505 in the air. These false positives disappear with the dual rider system. Areas of circles are structural objects (eg columns) that are present and detected in both tests.

5C and 5D demonstrate the use of a multi-plane lidar filter in detecting false positive echoes. FIG. 5C is a plot of a multi-plane rider (without filtering). The horizontal axis represents the distance from the front and rear surfaces of the vehicle. The vertical axis represents the distance from the side of the vehicle. Each segment represents a distance of 5 meters. Points along the horizontal axis indicate the detection of false positives due to rain.

The test was repeated using a (multi-planar) dual system in which the first rider system and the second rider system scan the area. FIG. 5D is a plot of a multi-plane rider using a dual rider system. Many of the plotted points disappear with the dual rider filter. These points are due to raindrops in the air. The points present in both FIGS. 5C and 5D can be due to physical objects (ie features).

Additional modifications / components can improve the robustness of the system during heavy rains (eg, rains above 20 mm / hour). One approach is to use a rider unit that can analyze additional layers (eg, 5 or more layers) to improve infiltration in the rain. Rider units with higher refresh rates may also be used to filter out raindrops. A mechanical air blower may be implemented to provide an air curtain with a range (eg up to 1 meter) to reduce the amount of raindrops detected by the rider unit.

It will be appreciated that the features and functions disclosed above and variations of other features and functions, or their alternatives, may be combined with other systems or applications. Also, various unexpected or unexpected substitutions, changes, modifications or improvements thereof, which are also covered by the claims below, may be made later by those skilled in the art.

Although embodiments of the present disclosure have been described in considerable detail and comprehensively to cover possible embodiments, one of ordinary skill in the art will recognize that other versions of the present disclosure are also possible.

Claims (17)

A system for distinguishing solids from particles in the air using laser light.
A first rider device that scans an area by emitting a pulse of laser light and detecting the reflected laser light.
A second rider device that scans the same or substantially the same area as the first rider device by emitting a pulse of laser light and detecting the reflected laser light.
From a computer that compares the pulse emitted by the first rider device and the detected reflected laser light to the pulse emitted by the second rider device and the detected reflected laser light. Configured
A system in which the presence of a solid object is confirmed when the first rider device and the second rider device detect laser light reflected from the solid object. The system of claim 1, wherein the first rider device and the second rider device utilize multi-echo technology to distinguish solids from particles in the air. The first rider device and the second rider device scan a plurality of planes to ensure that one or more echoes are reflected from solids rather than particles in the air. The system of claim 1, wherein echoes from one or more planes are compared. The system of claim 1, wherein the first rider device and the second rider device are located on a horizontal plane at a distance of 0.5 to 2.5 meters from each other. A method of distinguishing solids from particles in the air
A step of scanning an area close to an autonomous vehicle by emitting a pulse of laser light using a first rider unit and detecting an echo.
A step of scanning substantially the same area by emitting a pulse of laser light using a second lidar unit and detecting an echo.
A step of using an algorithm to compare the pulse emitted and received echo by the first rider unit with the pulse emitted and received echo by the second rider unit.
A method comprising a step of confirming the presence of the solid object when the first rider unit and the second rider unit detect an echo from the solid object. The method of claim 5, wherein the first rider unit and the second rider unit use multi-echo scanning. Claim that the first rider unit and the second rider unit scan a plurality of planes and compare echoes from one or more planes in order to distinguish a solid object from aerial particles. The method according to 5. The method of claim 7, wherein the presence of a solid is confirmed when echoes are detected in two or more planes. The method of claim 5, wherein the first rider unit and the second rider unit are located on a horizontal plane at a distance of 0.5 to 2.5 meters from each other. A method of distinguishing solids or obstacles from particles present in the air.
a) With the step of using the first rider device to scan the area by emitting a pulse of laser light and detecting the reflected laser light.
b) With the step of using a second rider device to scan substantially the same area by emitting a pulse of laser light and detecting the reflected laser light.
c) A step of collecting data related to the laser light emitted from the first rider device and the reflected laser light, and
d) A step of collecting data related to the laser light emitted from the second rider device and the reflected laser light, and
e) One of the first rider device and the second rider device emits reflected laser light that is not detected by both the first rider device and the second rider device. A method comprising the step of determining that the reflected laser light is the result of particles in the air when detected. 10. The method of claim 10, wherein the first rider device and the second rider device use multi-echo scanning. The first rider device and the second rider device scan multiple planes to distinguish solids from particles in the air and compare echoes from one or more planes. The method according to claim 10. 12. The method of claim 12, wherein the presence of a solid is confirmed when echoes are detected in two or more planes. It ’s a way to use a rider to check for the presence of solid objects.
a) The step of using the first rider device to scan multiple planes, where the laser light pulse is emitted and the reflected laser light is detected.
b) A step of collecting data relating to the laser light emitted from the first rider device and the reflected laser light detected by the first rider device.
c) With the step of determining whether the light reflected from the first rider device is the result of particles in the air by comparing multiple planes for the presence and / or absence of reflected light. ,
d) A step of using a second rider device to scan multiple planes where a pulse of laser light is emitted and the reflected laser light is detected.
e) A step of collecting data related to the laser light emitted from the second rider device and the reflected laser light detected by the second rider device.
f) By comparing multiple planes for the presence and / or absence of reflected laser light, it is determined whether the laser light reflected from the second rider device is the result of particles in the air. Steps and
g) Consists of a step of using an algorithm to compare the data from the first rider device with the data from the second rider device.
The plurality of planes scanned by the first rider device are in or substantially the same area as the plurality of planes scanned by the second rider device.
A method in which the presence of a solid object is confirmed when the first rider device and the second rider device detect laser light reflected from the solid object in a plurality of planes. Laser light reflected within four or more of the plurality of planes of the first rider device and / or four or more of the plurality of planes of the second rider device is detected. The method of claim 14, wherein the presence of a solid is confirmed when the laser is used. 14. The method of claim 14, wherein the first rider device and the second rider device use multi-echo scanning. 14. The method of claim 14, wherein the first rider device and the second rider device are located on a horizontal plane at a distance of 0.5 to 2.5 meters from each other.

Images