WO2020215745A1

WO2020215745A1 - Three-dimensional lidar and positioning method therefor

Info

Publication number: WO2020215745A1
Application number: PCT/CN2019/124732
Authority: WO
Inventors: 周逸铭; 周常站
Original assignee: 东莞市光劲光电有限公司; 周逸铭; 周常站
Priority date: 2019-04-22
Filing date: 2019-12-12
Publication date: 2020-10-29
Also published as: CN110231628A

Abstract

A three-dimensional lidar and a positioning method therefor, which relate to the field of lidars. The three-dimensional lidar comprises transmitting units (1), a receiving unit (2), a photoelectric inductor (3) and an image processor (4); the transmitting units (1) comprise laser tubes (11), and several laser tubes (11) are axially arranged outwards around the center of a circle to form an annular laser network; or the transmitting units (1) comprise laser tubes (11) and conical mirror bodies (13), the laser emission direction of the laser tubes (11) face conical surfaces of the conical mirror bodies (13), and the conical mirror bodies (13) reflect laser light to form an annular laser network; the receiving unit (2) is an optical structure of a positive cone of a mirror surface, receives diffusely reflected laser light and reflects same to the photoelectric inductor (3), and a three-dimensional point cloud image is calculated and processed by the image processor (4); and the laser tubes (11) are combined to make the laser network formed by the transmitting units (1) into an annular laser network, the diffusely reflected laser light generated by the annular laser network encountering surrounding obstacles is received by a receiver of the positive cone, and is reflected to the photoelectric inductor (3), and then image processing is performed. The foregoing three-dimensional lidar structure has a fast response speed and a large horizontal vision, and has a simple structure, low costs, and low production difficulty.

Description

Three-dimensional laser radar and its positioning method

Technical field

The invention relates to the field of laser radar, in particular to a three-dimensional laser radar and a positioning method thereof.

Background technique

With the rapid development of AI, more and more intelligent robots are used in large shopping malls, hospitals and factories to replace human work. At present, intelligent robots mainly rely on lidar for indoor positioning. Now solid-state lidar includes mechanical, MEMS mechanical/ There are three types of solid-state hybrid and phased array (matrix) lidars. At present, these radars have some shortcomings. For example, mechanical lidars are mainly driven by rotating motors. There are many laser tubes, short life, reliability, and slow response speed. ; MEMS-type solid-state hybrid radars have small horizontal angle scanning problems. Phased array solid-state lidars have complex structures, internal interference, and small horizontal scanning angles. Moreover, they all have high cost and high production difficulties.

Summary of the invention

In order to solve the above-mentioned problems, the present invention provides a three-dimensional laser radar and its positioning method. The laser net formed by the transmitting unit is a ring laser net through the combination of laser tubes, and the receiving unit simultaneously receives the diffusely reflected laser light and reflects it to the photoelectric sensor. Then image processing is performed. This three-dimensional lidar has fast response speed, large horizontal vision, simple structure, low cost and low production difficulty.

The technical scheme adopted by the present invention is: a three-dimensional laser radar, including a transmitting unit, a receiving unit, a photoelectric sensor, and an image processor; the transmitting unit is combined with a laser tube so that the laser network formed by the transmitting unit is a ring laser network;

There are two ways to make the laser net emitted by the transmitting unit a ring laser net;

The first method is: the emitting unit includes a laser tube, and a combination of several laser tubes emits to form a ring laser network.

The second way is that the emitting unit may also include a laser tube and a conical mirror body. The laser emitting direction of the laser tube faces the conical surface of the cone mirror body, and the cone mirror body reflects laser light to form a ring laser net.

In the above technical solution, when the emitting unit is formed by a plurality of laser tubes arranged outwards around the center of the circle to form a ring laser net, the laser tubes are arranged horizontally or inclined upward at a certain angle to form a horizontal ring laser net or a cone-shaped ring laser net .

In the above technical solution, the emitting unit has at least two layers, thus forming a three-dimensional laser network.

In the above technical solution, the receiving unit faces the transmitting unit to receive the laser light diffused by the obstacle after the transmitting unit emits; the receiving unit includes a receiving station and a spacer, and the spacers are evenly distributed on the receiving table to make the receiving unit Only the direct laser beam can accurately obtain the position of the diffused laser beam.

In the above technical solution, the receiving table is a right prism or truncated cone, and the side of the receiving table is a mirror surface to reflect the laser; the spacers are evenly distributed on the side of the receiving table's right prism or truncated cone, so that the direct laser beam can be accurate Reflected to the photoelectric sensor.

In the above technical solution, each emitting unit is electrically connected with an independent power source, so that different emitting units can be selected for combined positioning, and a two-dimensional space of any height can be scanned.

The present invention also adopts a three-dimensional lidar positioning method, the steps of which include:

A. At least two layers of ring laser nets are emitted by the emitting unit;

B. The laser encounters obstacles and diffuses to the receiving unit;

C. The receiving unit is reflected to the photoelectric sensor, and the image processor calculates the two-dimensional point cloud image diffused by each layer of the ring laser network, and combines it into a three-dimensional point cloud image.

In the above method, in step C, the calculation method for the image processor to calculate the two-dimensional point cloud image is the time-of-flight ranging method, and the distance and angle of the obstacle are obtained by the flight (round trip) time of the laser.

In the above method, in step C, the calculation method for the image processor to calculate the two-dimensional point cloud image may also be a triangulation ranging method, which is used to measure the distance and angle between the obstacle and the photoelectric sensor. .

In the above method, in step C, each layer of the ring laser network calculates a two-dimensional point cloud image, and at least two or more two-dimensional point cloud images are synthesized into a three-dimensional point cloud image to achieve the effect of scanning a two-dimensional space of any height.

The beneficial effects of the present invention are: a three-dimensional lidar and a positioning method thereof, the components used are simple, the production cost is low, the problem of high cost of the traditional lidar is solved, and the problem of complicated and difficult assembly and debugging of the original lidar is also solved. ; Through the laser tube combination, the laser network formed by the transmitting unit is a ring laser network, which can receive 360° horizontal laser signals, which solves the small problem of traditional laser radar horizontal viewing angle; and its positioning method can read all at once through photoelectric sensors The laser reflection signal of the obstacle is synchronously transmitted to the image processor for calculation and processing, which solves the problem of slow response speed of the lidar and has a wide range of use values.

Description of the drawings

Figure 1 is an overall front view of the three-dimensional lidar of the present invention;

2 is a schematic diagram of a transmitting unit according to the first embodiment of the present invention;

Fig. 3 is a schematic diagram of a transmitting unit according to the second embodiment of the present invention.

Fig. 4 is a schematic diagram of a transmitting unit according to the third embodiment of the present invention.

Fig. 5 is a schematic diagram of the receiving unit of the present invention.

Figure 6 is a schematic diagram of the laser route of the present invention.

Detailed ways

The present invention will be further described in detail below in conjunction with the drawings.

Example one:

Fig. 1, Fig. 2, Fig. 5 and Fig. 6 schematically show a three-dimensional laser radar and its positioning method according to an embodiment of the present invention.

A three-dimensional lidar, including:

The transmitting unit 1 includes at least two or more layers of transmitting units 1. The example here includes three layers of transmitting units 1, but the number of transmitting units 1 is not limited. The emitting unit 1 of each layer is placed at a different height in the space, and this height can be set by yourself. The emitting unit 1 can emit a circular laser beam horizontally and form a horizontal ring laser net, as shown by the dotted line in Figure 2; 1 Each ring laser can be emitted at different heights to form a three-dimensional ring laser net in space. Each layer of the emitting unit 1 includes a laser tube 11 and a laser circuit board 12. The laser circuit board 12 is electrically connected to the laser tube 11 to realize automatic control; a laser tube 11 is arranged horizontally along a circle, and the laser head of the laser tube 11 faces In addition, the tail and the axis of the laser tube 11 point to the center of the circle, and the horizontal angle between the axes of the adjacent laser tubes 11 is P=360°/a. The laser tube 11 can emit a linear laser line segment, and its emission line angle is also P= 360°/a, so that all the laser lines emitted by the laser tube 11 will be connected end to end, and finally form a horizontal ring laser net.

The receiving unit 2 is a ring-shaped laser receiving/reflecting device. The laser light emitted by the transmitting unit 1 will be diffusely reflected when it hits an obstacle. The receiving unit 2 can receive the diffusely reflected laser light and reflect it to the photoelectric sensor器3. The receiving unit 2 includes a receiving platform 21 and a spacer 22; the receiving platform is a right-edge platform, and the sides of the receiving platform 21 are electroplated to reflect mirror surfaces; the spacer 22 is made of light-absorbing material, and the spacer 22 is vertically installed on the side of the receiving platform 21 In this way, the entire side of the receiving platform 21 is isolated into n independent areas. Due to the existence of the spacer 22, each independent area can only receive light from the space directly in front of it, and light from other spaces will be The separator 22 shields and absorbs.

The photoelectric sensor 3 faces the reflective surface of the receiving platform 21 of the receiving unit 2 to receive the laser light reflected in the forward direction and accurately obtain positioning information. The photoelectric sensor 3 includes an avalanche diode array 31, an optical lens 32 and a control processing element; the optical lens 32 is arranged in front of the avalanche diode array 31, the optical lens 32 assists the avalanche diode array 31 to obtain laser signals, and the control processing element is electrically connected to the avalanche diode array 31. Obtain relevant data.

The image processor 4 is electrically connected to the photoelectric sensor 3 to realize the calculation of a two-dimensional point cloud image from the data obtained by the photoelectric sensor 3, which is then combined into a three-dimensional point cloud image to obtain the coordinates of obstacles and realize three-dimensional laser positioning.

As mentioned above, each layer of emitting unit 1 is electrically connected with an independent power source, so that each layer of emitting unit 1 is individually lit. Of course, in special cases, if a specific area needs to be scanned, the independent power supply can also light up the corresponding emitting unit 1 individually.

A three-dimensional laser radar positioning method, which applies the three-dimensional laser radar, and its steps include:

A. At least two layers of ring laser nets are emitted by the transmitting unit 1. The example here includes three layers of transmitting units 1, but the number of transmitting units 1 is not limited;

B. The laser encounters obstacles and diffuses to the receiving unit 2. When the laser diffuses to the receiving platform 21 of the receiving unit 2, if it is a forward laser line, it will be reflected to the photoelectric sensor 3, if it is not a forward The laser line will be absorbed by the spacer 22;

C. The receiving unit 2 is reflected to the photoelectric sensor 3, and the image processor 4 calculates the two-dimensional point cloud reflected by each layer of the ring laser net, and combines it into a three-dimensional point cloud.

As mentioned above, the calculation formula for the obstacle distance from the radar center point: Set a certain transmitting unit 1 to emit a circular laser at the initial time t, and the laser will diffuse and reflect diffusely when it hits the obstacle on the horizontal plane. The laser is received by the side of the receiving station opposite, and the received laser will be reflected to the corresponding area of the avalanche diode array 31 of the photoelectric sensor 3. In these areas, the avalanche diode 31 will produce avalanche phenomenon after receiving the laser photon and produce corresponding By recording the angle of the corresponding area and the start time t1, t2...tn of the electrical pulse signal, the distance L between obstacles in each area and the radar is calculated by the following formula. (The time consumed by the laser inside the radar is ignored)

The distance from the obstacle of angle 1 to the radar center point: L1=(t1-t)*C/2

The distance from the obstacle of angle 2 to the radar center point: L2=(t2-t)*C/2

The distance from the obstacle of angle n to the radar center point: Ln=(tn-t)*c/2

According to the angle and distance, a two-dimensional point cloud image is obtained, and then a three-dimensional point cloud image is obtained according to the height difference of each layer of the transmitting unit.

As mentioned above, each layer of the ring laser network calculates a two-dimensional point cloud image, and at least two or more two-dimensional point cloud images are combined into a three-dimensional point cloud image to achieve the effect of scanning a two-dimensional space at any height; of course, in special circumstances, if It is necessary to scan a specific area, or the corresponding transmitting unit 1 can be calculated separately.

Embodiment two:

Fig. 1, Fig. 3, Fig. 5 and Fig. 6 schematically show a three-dimensional laser radar and its positioning method according to an embodiment of the present invention.

A three-dimensional lidar, including:

The transmitting unit 1 includes at least two or more layers of transmitting units 1. The example here includes three layers of transmitting units 1, but the number of transmitting units 1 is not limited. The emitting unit 1 of each layer is placed at a different height in the space, and this height can be set by yourself. The emitting unit 1 can emit a circular laser beam horizontally and form a horizontal ring laser network, as shown by the dotted line in Figure 3; 1 Each ring laser can be emitted at different heights to form a three-dimensional ring laser net in space. Each layer of the emitting unit 1 includes a laser tube 11 and a laser circuit board 12. The laser circuit board 12 is electrically connected to the laser tube 11 to realize automatic control; the laser tube is inclined at an angle to form a conical arrangement, resulting in a conical shape Ring laser net.

The photoelectric sensor 3 faces the reflective surface of the receiving platform 21 of the receiving unit 2 to receive the laser light reflected in the forward direction and accurately obtain positioning information. The photoelectric sensor 3 includes a camera 34, an optical lens 32, and a control processing element; the optical lens 32 is arranged on the camera 34, the optical lens 32 assists the camera 34 to obtain laser signals, and the control processing element is electrically connected to the camera 34 to obtain related data.

Above, the calculation formula of the obstacle distance from the radar center point: As shown in Figure 6, a vertical plane is used at a certain angle in the horizontal direction, assuming this angle is ￡, ￡ is any value in 360°, and the entire When the 3D lidar is cut, you will get a cross-sectional view as shown in Figure 6. On the cross-sectional view, set the center point O(0,0) of the photosensitive surface of the camera 31 as the coordinate origin, the vertical direction as the y axis, and the horizontal direction as x Axis; set the laser emitting unit 11 to emit a ring laser, in this section it becomes a straight line VF, set the obstacle coordinate in this section as F, and the diffuse reflection of the laser light is by the ring receiver after the laser hits the obstacle Receive and project to the G point of the camera CMOS sensor, so that the length of OG can be tested by the camera, and the distance of OR (mirror distance) is also known, so we can calculate k by the formula tgk=OG/OR Because the angle k and the angle h are in a certain proportion, we can know the angle value of h;

Knowing the R coordinate value R(0,a) and h angle value, we can know the RW linear equation

y=tgh*x+a...Equation 1

At the same time, we also know the equation of the straight line XWE (because the angle i of the right edge table and the installation position are fixed as shown in Figure 5)

y=tgi*x+b...Equation 2

The intersection point of the two straight line equations 1 and the straight line equation 2 is the coordinate of point W(w1,w2)

From △REW, we know the angle q+h+[180-(90-i)]=180 q=-90-i-h

From △WFX we know the angle u+q+(180-i)=180 u=90+2i+h

Since both i and h are known, the angle value of u can be calculated,

Because the coordinates of W(w1,w2) are known, the angle of u is also calculated

We can know the WF linear equation y-w2=tgu*(x-w1)...Equation 3

The laser straight line VXF equation is y = c...equation 4

Calculate the coordinates of the intersection F of the straight line WF and the laser straight line VXF, which is the distance from the radar center point to the obstacle. In this example, the laser is a horizontal straight line. If the laser is an oblique line ZT (as shown in Figure 6), it can be calculated by the same principle.

Example three:

Fig. 1, Fig. 4, Fig. 5 and Fig. 6 schematically show a three-dimensional laser radar and its positioning method according to an embodiment of the present invention.

This kind of three-dimensional lidar is different from the three-dimensional lidar of the first and second embodiments in that each layer of the transmitting unit 1 includes a laser tube 11, a laser circuit board 12, and a conical mirror body 13, a laser circuit board 12 and a laser tube 11 is electrically connected to realize automatic control; the laser tube 11 is a point-shaped laser tube, the lower end of the conical mirror body 13 is a reflecting mirror with an inclined plane of 45°, and the conical mirror body 13 is vertically arranged directly above the laser tube 11. The emitting head of the tube 11 is directly opposite to the lower end of the reflector 13; the point-shaped laser emitted from the laser tube 11 hits the reflector surface of the cone lens body 13 and is reflected into a horizontal ring laser net, as shown by the dotted line in FIG. 4.

The above embodiments are only to illustrate rather than limit the present invention. If the above-mentioned scheme uses only one transmitting unit, a two-dimensional solid-state three-dimensional lidar point cloud image can also be generated separately, which is also protected by this patent. Therefore, according to the present invention Equivalent changes or modifications made by the methods described in the scope of the patent application are all included in the scope of the patent application of the present invention.

Claims

A three-dimensional lidar includes a transmitting unit, a receiving unit, a photoelectric sensor and an image processor; it is characterized in that,

The emitting unit includes a laser tube, and several laser tubes are arranged axially outwards around the center of the circle to form a ring laser net;

or,

The emitting unit includes a laser tube and a conical mirror body, the laser emitting direction of the laser tube faces the conical surface of the conical mirror body, and the cone mirror body reflects laser light to form a ring laser net.
The three-dimensional laser radar according to claim 1, wherein when the transmitting unit is formed by a plurality of laser tubes arranged axially outwards around the center of the circle to form a ring laser net, the laser tubes are arranged horizontally or inclined upward at a certain angle, A horizontal ring laser net or a conical ring laser net is formed.
The three-dimensional laser radar according to claim 1, wherein the transmitting unit has at least two layers to form a three-dimensional laser net.
The three-dimensional lidar according to claim 1, characterized in that: the receiving unit faces the transmitting unit to receive the laser light diffused by the transmitting unit; the receiving unit includes a receiving platform and a spacer, and the spacers are evenly distributed on the receiving unit. On the table surface, so that the receiving unit only receives the laser beam directly from the front, and accurately obtains the position of the diffused laser beam.
The three-dimensional lidar according to claim 4, characterized in that: the receiving platform is a right prism or a right truncated cone, and the side of the receiving platform is a mirror surface to reflect the laser; the spacers are evenly distributed on the prism or the right of the receiving platform. The side surface of the truncated cone enables the laser beam from the front to be accurately reflected to the photoelectric sensor.
A three-dimensional lidar according to claim 1, 2 or 3, wherein each transmitting unit is electrically connected with an independent power source, so that different transmitting units are selected for combined positioning, and can be used for a two-dimensional space at any height. To scan.
A three-dimensional lidar positioning method, the steps include:

A. At least two layers of ring laser nets are emitted by the emitting unit;

B. The laser encounters obstacles and diffuses to the receiving unit;

C. The receiving unit is reflected to the photoelectric sensor, and the image processor calculates the two-dimensional point cloud image diffused by each layer of the ring laser network, and combines it into a three-dimensional point cloud image.
The 3D lidar positioning method according to claim 7, characterized in that: in step C, the calculation method for the image processor to calculate the 2D point cloud image is a time-of-flight ranging method.
The 3D lidar positioning method according to claim 7, characterized in that: in step C, the calculation method for the image processor to calculate the 2D point cloud image is a triangulation ranging method.
A three-dimensional lidar positioning method according to claim 7, 8 or 9, characterized in that: in step C, a two-dimensional point cloud image is calculated for each layer of the ring laser network, and at least two or more two-dimensional point cloud images are synthesized Three-dimensional point cloud diagram.