(Original title: What can a single-line laser radar do in autopilot?) Definition of Lidar The earliest definition of laser radar is LIDAR, and English is Light Deteation and Ranging. Chinese means "light detection and distance measurement." In fact, a more accurate definition is LADAR: LAser Detection and Ranging, that is, "Laser Detection and Ranging." This is the definition put forward in 2004 and is more in line with the concept of laser radar. Lidar is actually a kind of radar that works in the optical band (special band). Its advantages are very obvious: 1, has a very high resolution: laser radar work in the optical band, the frequency of 2 to 3 orders of magnitude higher than the microwave, therefore, compared with the microwave radar, laser radar has a very high range resolution, angular resolution and speed Resolution; 2, strong anti-interference ability: laser wavelength is short, can emit a very small divergence angle (μrad level) of the laser beam, multipath effect is small (will not form a directional emission, and microwave or millimeter wave multipath effect), can be detected Low/ultra-low altitude targets; 3, to obtain a wealth of information: direct access to the target distance, angle, reflection intensity, speed and other information, generate target multi-dimensional images; 4, can work throughout the day: laser active detection, does not rely on external lighting conditions or the target's own radiation characteristics. It only needs to emit its own laser beam and acquire the target information by detecting the echo signal of the emitted laser beam. However, the biggest drawback of Lidar is that it is easily affected by the atmospheric conditions and the dust in the work environment. It is very difficult to achieve an all-weather working environment. Lidar classification Lidar classification, if divided from the system, mainly includes direct-detection laser radar and coherent detection laser radar. In fact, what we have mentioned at present, including autopilot, robotics, and laser radar used in surveying and mapping, is basically a direct-detection type of laser radar. There are special types of radars, such as wind measurement and speed measurement, which generally use a coherent system. According to the application classification, we can share more, such as: laser range finder, laser three-dimensional imaging radar, laser speed radar, laser atmospheric radar, and so on. Whether it is a single-line laser radar, a multi-line laser radar, or a mapping laser radar, we can basically divide it into the category of laser 3D imaging radars. A laser 3D imaging radar actually needs two core information: target distance information and target angle information. If we determine its three-dimensional sitting standard, we need to get its distance, azimuth, pitch angle information. Then we calculate the three-dimensional coordinates of the target based on the three information of distance, azimuth angle and pitch angle. In general, the technique of obtaining angle information by measuring an encoder is mature. What we are more concerned with is how the lidar distance information is obtained. The laser three-dimensional imaging radar can obtain the target three-dimensional point cloud data through direct distance measurement and direct angle measurement technology, and the obtained data itself is three-dimensional data, and does not require a large number of operations and processing to generate the target three-dimensional image, and the laser ranging has Very high accuracy. Therefore, the laser three-dimensional imaging radar is the most efficient sensor capable of acquiring images of a wide range of three-dimensional scenes at present, and is also the most accurate sensor capable of acquiring three-dimensional scenes at present. Laser ranging method At present, we can usually see the ranging method, which can be divided into the following categories: Time of Fly (TOF) and Triangulation. The laser flight time method can be divided into two types, one is pulse modulation (pulse ranging technology), and the other is the modulation of the intensity of the laser continuous wave, and the phase distance measurement of the distance information by the phase difference. The rangefinders we can see in the market, or single-line, multi-line laser radars, basically use these three types of ranging methods. Laser pulse distance measurement technology The principle of laser pulse distance measurement technology is very simple: to obtain the information of the target distance by measuring the time of flight of the laser pulse between the radar and the target. A benchmark is used here, which is the speed of light. All measurements must have a reference. For a laser, there are two references: speed and frequency (the two most accurate references) because the reference used for the TOF is the flight speed of the laser. The above-mentioned three kinds of distance measurement methods, I think the biggest technical difficulty is the pulse ranging method. But the advantages it brings are very clear: the measurement speed is very fast. Because of the measurement by the high-peak laser, its anti-glare interference ability is very strong. The disadvantage is that it is difficult to increase the resolution of the ranging, and the detection circuit is difficult. For example, if we want to achieve a phase resolution of 1.5 millimeters, we need to achieve a clock resolution of 10 picoseconds, which is equivalent to 100G bandwidth. This is a very difficult technique. Laser Phase Ranging Laser phase-ranging, such as common handheld laser rangefinders, uses phase-ranging methods. It mainly obtains the distance information by measuring the phase difference generated by the intensity modulated continuous wave laser signal between the radar and the target. The biggest advantage of this technology is that the range resolution is very high. At present, the phase distance meter on the market generally can achieve millimeter-level resolution. The disadvantage is that the measurement speed is slower than the pulse distance measurement. After all, we calibrate a phase difference to at least tens or even hundreds of cycles. In fact, it is equivalent to lengthening the measurement time, then its measurement speed. Relatively low. In addition, its measurement accuracy is relatively easily influenced by the target shape motion. If two targets are in front of and behind the measured spot, the actual information it measures is an average of the two target distances, not the previous target information or the latter target information. However, in pulse ranging, it is easy to separate such information. For example, a laser pulse, if we can achieve a pulse width of 10 nanoseconds, then we can separate a target by a multiple of echoes by a distance of 30 centimeters. This method is very difficult to distinguish it in phase ranging. Because in the measurement process, its time will be longer, the target movement brings in the distance information, it is introduced into the measurement value, in fact it measures an average distance information, not real-time information. However, the laser pulse distance measurement is actually the real-time information of the current position. This is also why laser radars for vehicles or robots often use laser pulse ranging technology instead of phase-ranging technology. Triangulation Triangulation distance measurement is to obtain distance information by measuring the imaging position of the laser irradiation point in the camera. The biggest advantage of the triangulation method is its low technical difficulty, low cost, and high accuracy at close range. For example, industrial use can achieve 100 micrometer ranging accuracy. However, the disadvantage is that its accuracy will gradually degrade as the distance increases, and it is basically impossible to compare with pulse ranging and phase ranging. Another point, because the CMOS camera must be illuminated with a continuous laser synchronization, its average power is relatively low, and the anti-jamming capability is very strong. This kind of distance measurement method is generally suitable for indoor close working, not suitable for Outdoor glare background or work with indoor glare background. Triangulation range measurement is more suitable for robots and other applications that do not require high performance. From the above figure, we can see that pulse ranging is superior to other aspects in terms of cost and technical difficulty. Of course, its ranging accuracy will be slightly lower than the accuracy of phase-ranging. However, with this accuracy, according to the current technology, we can basically achieve centimeter-level or even millimeter-level ranging accuracy, which can basically meet the requirements of our many applications. Our main direction is to use pulse ranging to do single-line radars, including multi-line radars. What is a single line laser radar At present, single-line laser radar products mainly include SICK Corporation and HOKUYO Corporation. The single-line laser radar is actually a high-frequency pulse laser rangefinder plus a one-dimensional rotation scan. Single-line laser radar features: 1, only one way to launch and one way to receive, the structure is relatively simple, easy to use; 2, high scanning speed, high angle resolution; 3, low volume, weight and power consumption; 4, higher reliability; 5, low cost; What can a single line laser radar do? In the field of automated driving, what we basically see is a multi-line laser radar. What can a single-line laser radar do in the end? As shown above, in the US DARPA Autodriving Challenge, the first one was the 2005 Stanford University named Stanly, which was the championship car that year. The other is the car at Carnegie Mellon. At that time, they were basically using SICK's single-line laser radar. In particular, the Stanford University's car is equipped with five laser radars just above it. We can think of it as the "original ancestor" of the multi-line laser radar, but it uses five single-line laser radars to implement multi-line laser radars. The function. After Velodyne introduced the 64-line laser radar in 2007, many autopilot vehicles basically use Velodyne's products. But does this mean that there is no market for single-line laser radar in auxiliary or automatic driving? I don't think so. Because the single-line laser radar has its characteristics, for example, in the high repetition rate, high-angle resolution, multi-line laser radar is difficult to achieve the same technical indicators. The picture above is a photograph of some cars participating in this year's China Smart Car Future Challenge in Changshu. It can be seen that in addition to using Velodyne multi-line laser radar and Ibeo multi-line laser radar, they all installed SICK single-line laser radar. Xi'an Jiaotong University used our single-line laser radar because of its cooperation with us. In terms of pedestrian detection, obstacle detection (small target detection), and detection of obstacles ahead, single-line lasers are much more advantageous than multi-line laser radars because the angular resolution of single-line laser radar can be made higher than that of multi-line laser radars. This is very useful for detecting small objects or pedestrians. The image above is a screenshot of the propaganda video of Ibeo's single-line laser radar at that time. One of the functions it implemented is the forward collision prevention. For example, when a car suddenly appears on the left side of the road, the lidar can detect the location and size of the car and automatically avoid it. This technology is very useful in today's intelligent robots and service robots, and it is currently a hot area. The above figure shows the application of lidar in lane detection and lane yaw warning. Many people may ask a question: Why use LADAR to do lane detection without using cameras? The ADAS algorithm is not very mature. Why not use LIDAR? This is because the camera is particularly vulnerable to background light or strong light. For example, when we walk along the boulevard, if the shade of trees falls behind the spotted sunlight and then combined with the white lane line, it is very difficult for us to identify the lane line, and the recognition probability is under complex illumination or under strong light conditions. Its recognition probability is very low and the algorithm is very complicated. So, what are the advantages of using Lidar for lane detection? First, we use an infrared laser, which itself emits much lower infrared radiation than visible light. Second, we will add a very narrow filter to filter out strong background light directly. Then we use infrared light to detect it. In this way, we can obtain a very high-quality lane line image. Through the grayscale of the image, we can easily detect the lane line. In other words, using laser radar to do lane detection, its performance will be higher than the camera. The application of single-line laser radar in assisted driving is pedestrian detection. In fact, this is also a forward anti-collision application, which is basically similar to anti-collision of automobiles. Because the angular resolution of a single-line laser radar can be higher than that of a multi-line laser radar, pedestrians can be found ahead of time to provide more warning time for the control system or the driver. ACC (Stop & Go) application. This function is particularly applicable in the current situation of traffic jams in Chinese cities. It uses the forward laser radar to directly detect the previous car movement to obtain the precise distance information of the preceding vehicle, and then automatically follows the control car. Wonderful question and answer Q: Can I use multiple lines to simulate the minor lines? For example, if you select 16 lines of data from 64 lines of data, what is the difference between using 16 lines of laser data directly? Zhu Shaohao: It is possible to simulate many lines with multiple lines. The Velodyne multi-line laser radar can be designed. However, if you select a few lines from multiple lines, the resolution in the pitch angle will decrease. Because there is no fixed angle between the lines of the multi-line Lidar, either the interval selection or the continuous selection. Mainly based on actual needs to choose. Q: For solid-state lidars, how much difference will there be in terms of technical difficulty in manufacturing single-wire and multi-wires? Zhu Shaojun: If it is a Lidar that can perform two-dimensional scanning solid-state scanning, the difficulty of manufacturing single-line and multi-line is not very different. However, if the solid-state scanning device can only achieve one-dimensional scanning, multi-line radars must use multiple single-line radars to fight. It is necessary to solve the problem of coordination and crosstalk between each module, and the volume will be large. Lidar will be more difficult. Q: For light-absorbing obstacles, there is no guarantee that light will be effectively reflected back. At present, does Lidar technology have corresponding measures to solve this problem? Zhu Shaojun: This is a dead zone. If the laser radar encounters a light-absorbing obstacle, it will not be able to effectively receive the echo signal reflected from the target and will not be able to measure the position information of the target. If it is a mirror reflection obstacle, it will form a multipath effect. If you encounter a transparent glass, it is possible to detect the target behind the glass. This is determined by the physical properties of light. To solve the problem of light-absorbing obstacles, from my current knowledge, it can only be solved by increasing the power of laser emission, but it is not a good method. Q: Is the current single-line, multi-line laser radar program a transitional solution that will eventually transition to the area array lidar without a rotating mechanism that can be integrated with the body without the overhead angle? Zhu Shaojun: Yes. I think Lidar certainly will be scanned by the scanning radar to the gaze, that is, the development of the area imaging lidar, and even want to develop the direction of multi-spectral imaging, and at the same time can obtain the target spectral information. Once gaze imaging is achieved, Lidar is as lightweight as today's visible light cameras and can be perfectly integrated with the body. However, the current laser radar's area array detector is very difficult and costly. Q: Does the light source chip detector chip of the area array laser radar have a better solution or product in China? What do you know about the institutions that conduct this research? What are the research institutions that are likely to have breakthroughs in this area? Zhu Shaojun: At least four research institutes of area array lidars are currently engaged in research and development in this area. They are institutes for weapons, electricity, and sciences, including our Xi'an Institute of Opto-Electronics, but not yet. See commercial products come out. Because I am not very familiar with the technical status of other institutes, I cannot make accurate judgments. Samples of Xi'an Optics and Optical Machines have already come out, but the performance needs further optimization. Q: How many lasers per second do the HLD-1L01 Laser Point have? The TOF was also mentioned in the previous sharing. Is TOF not very weak outdoors? Zhu Shaojun: The current HLD-L01 of Aozhi Zhihua is currently 36k points/s. The TOF method using the pulse distance measurement system has strong anti-glare interference and is relatively low in outdoor exposure. In our road test image, the vehicle was traveling westward at 5 o'clock in the afternoon and its performance was not affected under strong light conditions. Q: Currently there is no laser radar industry chain in China? Lidar's algorithm is generally provided by who? What is the price of a single line lidar and the price curve for the next five years? Zhu Shaojun: At present, the domestic laser industry chain is mature. Lidar algorithm can be provided by many scientific research institutes and companies. For example, our partner Hongyi Yida can provide algorithms, and our company is currently doing some of this development work. The price of a single-line laser radar is mainly related to the production volume. The price curve for the next five years is not currently predictable. If the amount does not go, the cost will certainly not be too low. 120T Injection Machine /rotary Machine silicone injection machine,price of liquid silicone rubber injection molding machine,Liquid injection molding,Liquid Silicone filling machine Guangzhou S-guangyu Machinery &Equipment Co.,Ltd , https://www.lsrexpert-better.com