Method and apparatus for detecting objects using light pulses with non-uniform power
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HUAWEI TECH CO LTD
- Filing Date
- 2021-05-31
- Publication Date
- 2026-06-26
Smart Images
Figure CN115667992B_ABST
Abstract
Description
[0001] Cross-references
[0002] This application claims the benefit of priority to U.S. non-provisional application No. 16 / 889,582, filed on June 1, 2020, entitled “Method and Apparatus for Detecting an Object Using Optical Pulses with Non-uniform Power”. Technical Field
[0003] This invention relates to object scanning schemes, and more particularly to a method and apparatus for detecting objects within a field of view using light pulses with non-uniform pulse power. Background Technology
[0004] Advanced Driver Assistance Systems (ADAS) are electronic systems that assist drivers while driving or parking. These systems have been developed in the automotive industry to automate, adjust, or enhance vehicle systems for safer and easier driving. With the development of autonomous vehicles, ADAS has received increasing attention from the automotive industry. ADAS uses a set of sensors to detect and classify various objects within a defined field of view (FOV). One type of sensor used in ADAS is the light detection and ranging (LiDAR) sensor.
[0005] Time-of-flight (TOF) LiDAR sensors emit laser pulses and receive light pulses reflected or backscattered from a target object. Therefore, TOF LiDAR can calculate the distance to the target object based on the time difference between the emitted and received light pulses.
[0006] According to the TOF LiDAR equations, the power of a light pulse returning from or backscattering from a target object is inversely proportional to the square of the distance to the target object. In fact, detecting distant objects (e.g., objects far from the sensor) is more challenging due to the low signal-to-noise ratio (SNR). Therefore, increasing the power of the light pulse is desirable to improve the SNR, especially when detecting distant objects. However, for eye safety considerations, the power of the laser pulse emitted from a TOF LiDAR cannot exceed a certain power level (e.g., the admissible exposure limit (AEL)). Therefore, a method and apparatus are needed to detect objects (especially distant objects) within a defined field of view (FOV) that are not subject to the above limitations.
[0007] The purpose of the background art is to disclose information that the applicant believes may be relevant to the present invention. It is neither necessary nor appropriate to acknowledge that any of the foregoing information constitutes prior art in relation to the present invention. Summary of the Invention
[0008] The purpose of this invention is to provide a method and apparatus for detecting objects using light pulses with non-uniform power. According to an embodiment of the invention, a method for detecting objects within a field of view (FOV) is provided. The method includes emitting a first set of light pulses, wherein the first set of light pulses includes a first light pulse having a first pulse power and a second light pulse having a second pulse power. The first pulse power differs from the second pulse power, and the power of each pulse in the first set of light pulses is less than or equal to a predefined limit. The method further includes: receiving a first set of returned light pulses, wherein the first set of returned light pulses represents information related to the object within the FOV; and generating the information related to the object within the FOV from the first set of returned light pulses.
[0009] According to an embodiment of the present invention, an apparatus for detecting objects within a field of view (FOV) is provided. The apparatus includes: one or more light pulse emitters for emitting a plurality of light pulses, wherein the plurality of light pulses includes a first light pulse having a first pulse power and a second light pulse having a second pulse power. The first pulse power is different from the second pulse power, and the power of each pulse of the plurality of light pulses is less than or equal to a predefined limit. The apparatus further includes: one or more light pulse receivers for receiving a plurality of returned light pulses, wherein the plurality of returned light pulses represent information related to the object within the FOV. The apparatus further includes: a processor for generating the information related to the object within the FOV from the plurality of returned light pulses.
[0010] Embodiments have been described above in conjunction with various aspects of the present invention, and these embodiments can be implemented based on these aspects. Those skilled in the art will understand that embodiments can be implemented in conjunction with the aspects described therein, but may also be implemented together with other embodiments of that aspect. It will be apparent to those skilled in the art that embodiments are mutually exclusive or incompatible with each other. Some embodiments may be described in conjunction with one aspect, but may also be applicable to other aspects, as will be apparent to those skilled in the art.
[0011] Some aspects and embodiments of the present invention can improve the detection probability of light pulses returned from distant objects within the field of view (FOV). Increasing the detection probability of returned light pulses increases the point cloud density of distant objects, thus expanding the range of objects that can be detected within the FOV. Embodiments of the present invention also improve and expand the detection probability and range of objects with low reflectivity or in adverse weather conditions (e.g., fog, rain). Attached Figure Description
[0012] Further features and advantages of the invention will become apparent from the following detailed description taken in conjunction with the accompanying drawings.
[0013] Figure 1 This demonstrates the basic principles of time-of-flight (TOF) light detection and ranging (LiDAR).
[0014] Figure 2A and Figure 2B This shows how the human eye is exposed to multiple light pulses when LiDAR scans a target object.
[0015] Figure 2C The accessible emission limit (AEL) and C6=1 for Class 1 and Class 1M laser products in the Australian / New Zealand standard are shown.
[0016] Figure 2D The correction factors and transition points used in the assessment of AEL and maximum permissible exposure (MPE) in the Australian / New Zealand standards are shown.
[0017] Figure 3 The probability of light pulse detection under different signal-to-noise ratios (SNR) is shown.
[0018] Figure 4The image shows a point cloud generated by an optical pulse detector for a distant object using existing technology.
[0019] Figure 5 This invention illustrates a method for detecting distant objects within a field of view (FOV) according to an embodiment of the present invention.
[0020] Figure 5A A typical emission scheme for optical pulse detectors in the prior art is shown.
[0021] Figures 5B to 5D The present invention illustrates an emission scheme for a light pulse detector that uses light pulses with non-uniform power to detect objects, according to an embodiment of the present invention.
[0022] Figure 6 The embodiments of the present invention provide the object detection probabilities evaluated for uniform pulse power schemes and non-uniform pulse power schemes under different signal-to-noise ratios (SNR).
[0023] Figure 7 The illustration shows point clouds generated for distant objects using conventional uniform pulse power schemes and non-uniform pulse power schemes, as provided in an embodiment of the present invention.
[0024] Figure 8 This is a numerical analysis of the point cloud density at different distances using the traditional uniform pulse power scheme and the non-uniform pulse power scheme provided in the embodiments of the present invention.
[0025] Figure 9 The embodiments of the present invention provide the object detection probabilities evaluated for uniform pulse power schemes and non-uniform pulse power schemes under different signal-to-noise ratios (SNR).
[0026] Figure 10 This is a schematic diagram of an electronic device provided in an embodiment of the present invention.
[0027] It should be noted that similar features are identified by similar reference numerals throughout the accompanying drawings. Detailed Implementation
[0028] The term "AEL" refers to the Accessible Emitted Limit, which is obtained through a complex function that can include variables such as wavelength, repetition rate, pulse width, or pulse length. In this invention, the AEL is calculated according to the standard published by the Australian / New Zealand Joint Standards in "Safety of laser products Part 1: Equipment classification and requirements" at the Australian Standards Bureau / New Zealand Standards Organization meeting in Sydney in 2014.
[0029] This invention describes a method and apparatus for detecting objects (especially distant objects within the field of view (FOV)) using light pulses with non-uniform pulse power but not exceeding a predefined limit (e.g., AEL). According to embodiments of the invention, the probability of detecting light pulses returning from distant objects is increased. This increases the point cloud density of distant objects, particularly compared to existing technologies. In various embodiments, such distant objects can be detected using light detection and ranging (LiDAR) (e.g., time-of-flight (TOF) LiDAR).
[0030] According to an embodiment, a method for detecting a distant object within a field of view (FOV) is provided. The method includes: emitting a first set of light pulses having a non-uniform pulse power. This first set of light pulses may be emitted by a light pulse emitter included in a device for detecting an object within the FOV. The method further includes: receiving a first set of returned light pulses returned in response to interaction with the object. The first set of returned light pulses may represent information related to the object within the FOV. The method further includes: generating the information related to the object within the FOV from the first set of returned light pulses. In some embodiments, the method further includes: emitting a second set of light pulses having a non-uniform pulse power; and receiving a second set of returned light pulses. The second set of returned light pulses may also represent information related to the object within the FOV.
[0031] Figure 1 This illustrates the basic principles of time-of-flight (TOF) light detection and ranging (LiDAR). The distance to the target object 130 (i.e., d) is... 目标The time difference between the time when the transmitter 110 emits a light pulse (e.g., a laser pulse) and the time when the receiver 120 receives the light pulse returned or backscattered from the target object 130 can be estimated.
[0032] The light pulse returned from target object 130 needs to be received by receiver 120 in order to estimate the distance between LiDAR 100 and target object 130. Therefore, the power of the light pulse returned from target object 130 must also be strong enough to reach receiver 120. The power P of the returned light pulse... rx (t) the distance from LiDAR 100 to the target object 130 (i.e., d) 目标 It is inversely proportional to the power P of the returned light pulse. rx (t) and the power P of the emitted light pulse tx (t) is directly proportional. This can be expressed mathematically as:
[0033]
[0034] As shown in the mathematical expression above, the signal-to-noise ratio (SNR) can be significantly attenuated, especially for distant objects. On the other hand, SNR can also be increased by increasing the power P of the emitted light pulse. tx (t) to improve. As described elsewhere in this document, for eye safety considerations, the power P of the emitted light pulse is... tx The increase in (t) is limited by AEL.
[0035] Figure 2A This illustrates how the human eye 200 is exposed to multiple light pulses 250 when LiDAR scans a target object. For eye safety considerations, the total energy (or total power) of the light pulses 250 received by the human eye 200 is limited by AEL (Advanced Electron Emission Level). 总能量 At least in some embodiments, AEL 总能量 According to Figure 2C calculate. Figure 2C The achievable emission limits and C6=1 for Class 1 and Class 1M laser products in the Australian / New Zealand standard are shown. Figure 2C It is a copy of Table 3 provided by the Australian / New Zealand Standards. (Reference) Figure 2C AEL 总能量 The duration of light pulse emission (T) is determined based on the emission duration of the light pulse. The emission duration (T) of the light pulse can be determined as follows:
[0036] T = N / f
[0037] Where N is the number of light pulses received by the human eye in each scan, and f is the repetition rate of the scan light pulses, in Hertz (Hz).
[0038] According to existing technology, the energy per pulse (AEL) 每脉冲 It can be obtained by averaging AEL 总能量 To estimate, as shown below:
[0039] AEL 每脉冲 =AEL 总能量 / N
[0040] Where N is the number of light pulses received by the human eye during each scan.
[0041] Figure 2B The simulation estimates AEL are shown. 总能量 AEL 每脉冲 The number of pulses received by the human eye from an optical pulse detector (e.g., LiDAR 210). The specifications of the optical pulse detector (LiDAR 210) used in the simulation are shown in Table 1 below.
[0042] Table 1
[0043] Repetition rate 100kHz Horizontal angular resolution 0.1° laser wavelength 905nm Pulse width 5ns
[0044] To determine the number of pulses N exposed to the human eye at 200° per scan, the exposure angle must first be determined. (Reference) Figure 2B The exposure angle α240 can be estimated based on the length d230 and the distance D220 between LiDAR 210 and the human eye 200, as shown below:
[0045]
[0046] It should be noted that distance D 220 can refer to the minimum possible distance between LiDAR 210 and the human eye 200. Distance D 220 can be defined by eye safety standards (e.g., nationally defined standards, such as the Australian / New Zealand standard). It is easy to understand that other standards can be used to define these parameters. It should also be noted that length d 230 represents the diameter of the pupil of the human eye 200. Once the exposure angle α 240 is determined, the number of light pulses exposed to the eye 200 can be calculated based on the horizontal angular resolution of LiDAR 210, as follows:
[0047]
[0048] Given the repetition rate (f) of the light pulses emitted by LiDAR 210 (e.g., f = 100 kHz), in order to determine the AEL 总能量The total emission duration (T) of the optical pulses from LiDAR 210 needs to be calculated. It should be noted that the repetition rate (f) can be any other value readily understood by those skilled in the art. Furthermore, assuming the total emission duration (T) of the optical pulses can be calculated using T = N / f, then the total emission duration T of the optical pulses is equal to 330 μs (i.e., 33 pulses / 100 kHz = 330 μs).
[0049] Given the total emission duration T of the optical pulse, AEL 总能量 According to Figure 2C Confirmed. According to... Figure 2C Assume the laser wavelength is 905 nm (note: 905 nm is in the range of 700 nm to 1050 nm) and the total emission time is 330 μs (note: 330 μs is in the range of 1.3 × 10⁻⁶ nm). –5 s~1×10 –3 Within the range of s, then AEL 总能量 Equals 7 × 10 –4 t 0.75 C4 J (joules). According to Figure 2D Parameter C4 can be determined based on the wavelength of the light pulse. Figure 2D The correction factors and breakpoints used in the AEL and maximum permissible exposure (MPE) assessments in the Australian / New Zealand standards are shown. Figure 2D It is a copy of Table 9 provided by the Australian / New Zealand Standards.
[0050] according to Figure 2D When the laser wavelength is 905nm (note: 905nm is in the range of 700nm to 1050nm), parameter C4 equals 10. 0.002(λ–700) Therefore, assuming the laser wavelength is 905 nm (i.e., λ = 905 nm), then parameter C4 equals 2.57 (Note: 100.002). (905–700) =2.57).
[0051] AEL can be calculated based on the determined value of parameter C4. 总能量 As shown below:
[0052] AEL 总能量 =7×10 -4 ×(330×10 -6 ) 0.75 ×2.57=4.4μJ
[0053] According to existing technology, the energy per pulse (AEL) 每脉冲 It can be obtained by averaging AEL 总能量 To estimate, as shown below:
[0054]
[0055] Therefore, according to existing techniques using uniform pulse power levels, the energy per pulse is limited to 133.5 nJ (i.e., 26.7 watts of peak power), where AEL 总能量 Equal to 4.4 μJ, the specifications of the optical pulse detector (e.g., LiDAR) are shown in Table 1 above.
[0056] Figure 3 The probability of light pulse detection at different signal-to-noise ratios (SNR) is shown when estimating using a light pulse detector (i.e., a Neyman Pearson detector). Figure 3 This indicates that when the SNR is low, especially when the SNR is below 5dB, the probability of pulse detection is extremely low.
[0057] Since pulse power is inversely proportional to the distance to the target object, while SNR is directly proportional to pulse power, distant objects have lower SNR. Therefore, considering that the probability of light pulse detection is lower when SNR is low, the quality (e.g., density) of the point cloud provided by TOF LiDAR deteriorates significantly for long-distance objects. This clearly demonstrates that optical detectors (such as TOF LiDAR) are very likely to fail to detect (or not detect at all) distant objects correctly.
[0058] Figure 4 This illustrates a point cloud generated by an optical pulse detector for a distant object using existing techniques. (Reference) Figure 4 Train 400 has very little point cloud (i.e., low point cloud density), while street 450 has high point cloud density. Since train 400 is an object that needs to be detected more significantly than street 450 (e.g., for ADAS implementation), a low point cloud density for train 400 is undesirable. Since the power of the light pulse is proportional to the detection probability of the target object, and the detection probability of the target object is proportional to the point cloud density of the target object, increasing the power of the emitted light pulse could increase the point cloud density. However, considering that the light pulses emitted by light pulse detectors (e.g., TOF LiDAR) have a uniform power level, it is unlikely that the power of the light pulse can be increased due to AEL limitations, and therefore it is also unlikely that the point cloud density of distant objects can be increased using current techniques.
[0059] Figure 5 This invention illustrates a method for detecting distant objects within a field of view (FOV) according to an embodiment of the present invention. (Reference) Figure 5A method 500 is provided for detecting distant objects within a field of view (FOV) using one or more sets of optical pulses (e.g., laser pulses) with non-uniform optical pulse power. The power per pulse of the first set of optical pulses is less than or equal to a predefined limit, such as the average efficiency per pulse (AEL) (e.g., the average of the total AEL). In some embodiments, in step 501, an optical pulse generator may generate the first set of optical pulses (e.g., laser pulses). The generated optical pulses have non-uniform pulse power. For example, the first set of optical pulses may include one or more optical pulses with a first optical pulse power and one or more other optical pulses with a second optical pulse power. The first optical pulse power and the second optical pulse power are different from each other. The first optical pulse power and the second optical pulse power may be adjustable.
[0060] In some embodiments, the light pulse generator is included in the device for detecting objects within the FOV. In some other embodiments, the light pulse generator is not included in the device for detecting objects within the FOV, but is an external light pulse generator.
[0061] According to an embodiment, when a set of light pulses with non-uniform pulse power is generated in step 501, in step 502, one or more light pulse emitters can emit this generated set of light pulses. The one or more light pulse emitters can be part of a device for detecting objects within the FOV. In step 503, one or more light pulse receivers can receive a set of light pulses returned or backscattered from an object (e.g., a target object) within the FOV. The one or more light pulse receivers can be part of a device for detecting objects within the FOV. According to an embodiment, this set of returned light pulses (e.g., a set of light pulses returned from a target object) can represent information related to the object within the FOV. Once this set of returned light pulses is received, in step 504, information related to the object within the FOV can be generated from the returned light pulses. The object-related information can include three-dimensional information, such as a point cloud representing the object within the FOV. In some embodiments, the generated information can be transmitted to other external devices, but this is within... Figure 5 Not shown in the image.
[0062] According to some embodiments, steps 505 to 508 can optionally be performed in a manner similar to or the same as steps 501 to 504. Optional steps 505 to 508 can improve the detection probability of distant target objects within the FOV.
[0063] Now for further reference Figure 5In step 505, in addition to the first set of optical pulses generated in step 501, the optical pulse generator can also generate a second set of optical pulses (e.g., laser pulses). The second set of optical pulses generated in step 505 can also have non-uniform pulse power. The power per pulse of the second set of optical pulses is less than or equal to a predefined limit, such as the average efficiency per pulse (AEL) (e.g., the average of the total AEL). Those skilled in the art will understand that the second set of optical pulses can be different from the first set of optical pulses generated in step 501. For example, the second set of optical pulses may include one or more optical pulses with a third optical pulse power and one or more other optical pulses with a fourth optical pulse power. The third and fourth optical pulse powers are different from each other. However, the third or fourth optical pulse power can be the same as one of the first or second optical pulse powers. The third and fourth optical pulse powers can be adjustable.
[0064] In some embodiments, the light pulse generator is included in the device for detecting objects within the FOV. In some other embodiments, the light pulse generator is not included in the device for detecting objects within the FOV, but is an external light pulse generator.
[0065] According to an embodiment, when a second set of light pulses is generated in step 505, in step 506, one or more light pulse emitters can emit the second set of light pulses, as in step 502. In step 507, a light pulse receiver can receive a set of light pulses returned or backscattered from an object (e.g., a target object) within the FOV, as in step 503. According to an embodiment, the second set of returned light pulses (e.g., a second set of light pulses returned from the target object) can also represent information related to the object within the FOV. Once this set of returned light pulses is received, in step 508, information related to the object within the FOV can be generated from the second set of returned light pulses. The object-related information may include three-dimensional information, such as a point cloud representing the object within the FOV. In some embodiments, the generated information can be transmitted to other external devices, but this is within... Figure 5 Not shown in the image.
[0066] Figure 5A A typical emission scheme for optical pulse detectors in the prior art is shown. (Reference) Figure 5A Each square (i.e., each pixel 510) represents a scan pixel of a conventional optical pulse detector that emits light pulses with a uniform pulse power level (pulse energy). The pulse power of the emitted light pulse can be equal to the average value of the total AEL (i.e., AEL). 每脉冲 =AEL 总能量 / N). In other words, in the prior art, the average pulse power of each pulse is equal to AEL. 每脉冲 =AEL 总能量 / N.
[0067] On the other hand, according to an embodiment, in contrast to the prior art, a new method for detecting objects uses light pulses with non-uniform power. Figures 5B to 5D The emission scheme of the light pulse detector used in the novel method for detecting objects using light pulses with non-uniform power provided by an embodiment of the present invention is illustrated.
[0068] refer to Figure 5B The squares located in the leftmost, center, and rightmost columns are pixel 520, and the squares located in the remaining columns (i.e., the second leftmost and second rightmost columns) are pixel 530. Pixel 520 indicates that the pulse power (i.e., pulse energy) of the emitted light pulse is greater than... Figure 5A The scanning pixels of the optical pulse detector used have a uniform pulse power level (i.e., uniform energy level). It should be noted that... Figure 5A The pulse power level is equal to the average value of the total AEL (i.e., AEL). 每脉冲 =AEL 总能量 / N). On the other hand, pixel 530 indicates that the pulse power (i.e., pulse energy) of the emitted light pulse is less than Figure 5A The scanning pixels of the optical pulse detector used have a uniform pulse power level (i.e., uniform energy level).
[0069] According to an embodiment, the new method uses two or more pulse power levels for pixels within the field of view. The two or more pulse power levels (i.e., non-uniform pulse power) of the emitted light pulses are subject to the following conditions:
[0070] a. AEL of non-uniform pulse power 总能量 AEL equal to uniform pulse power 总能量 (i.e. AEL) 总能量 constant);
[0071] b. Maximum pulse power is limited by AEL during pulse time. 每脉冲 .
[0072] It should be noted that, except Figure 5B In addition to the pulse power distribution modes provided, other modes can be used in various embodiments. For example, Figure 5C and Figure 5D Different power schemes (e.g., pulse power distribution patterns) for use in optical pulse detectors are provided. According to embodiments, new methods for detecting objects can employ various schemes using optical pulses with non-uniform power.
[0073] It should also be noted that, although only an embodiment with two pulse power levels is shown in this invention for simplicity, more than two pulse power levels can be used in other embodiments as long as the above constraints are met.
[0074] According to embodiments, the optical pulse detector can scan the field of view (FOV) more than once, and multiple power schemes (e.g., pulse power distribution modes) can be used during FOV scanning. For example, in the first scan, the emitter of the optical pulse detector can use... Figure 5B The power scheme is used to emit light pulses. Then, in the second scan, the emitter of the light pulse detector can use... Figure 5B The power scheme shown is the opposite mode for emitting light pulses. Alternatively, in the second scan, the emitter of the light pulse detector can use... Figure 5C or Figure 5D The non-uniform power scheme shown is used to emit light pulses. Different modes can also be used when scanning the FOV (e.g., Figure 5D The non-uniform power scheme is obtained by a displacement pattern in the model.
[0075] As described above, in various embodiments, depending on the power scheme used to detect the object, two or more pulse power levels can be integrated with the optical pulse. For example, two pulse power levels can be integrated with the optical pulse, and with... Figures 5B to 5D The same applies to the binary power scheme. AEL for non-uniform pulse power (i.e., pulse power with two pulse power levels). 总能量 AEL equal to uniform pulse power 总能量 (i.e. AEL) 总能量 (Unchanged). Furthermore, the maximum pulse power per pulse is limited by the AEL within the pulse time. 每脉冲 .
[0076] The calculation of the two pulse power levels is further illustrated through a specific simulation example. For the simulation, it is assumed that a light pulse detector with the specifications provided in Table 1 above is used.
[0077] According to the embodiment, assuming the optical pulse detector has the specifications in Table 1, the maximum pulse power level of a single pulse can be determined using the pulse width. Since the pulse width is equal to 5 ns in the simulation example, the emission time of the optical pulse is at least 5 ns. Considering this, the AEL is determined... 每脉冲 To limit the maximum pulse power level of a single pulse, the emission time is equal to 5 ns, because the required minimum emission time (i.e., pulse width) is 5 ns. Assuming the laser wavelength is 905 nm (i.e., 905 nm is in the 700 nm–1050 nm range) and the emission time is 5 ns (i.e., 5 ns is in the 10… –9 s~10 –7 Within the range of s), then according to Figure 2C The maximum AEL of a single pulse is 7.7 × 10⁻⁶. –8C4 J (joules). Assuming the laser wavelength is 905nm (i.e., 700nm < λ = 905nm < 1050nm), then according to... Figure 2D The parameter C4 equals 2.57 (i.e., 10). 0.002(λ–700) =10 0.002(905–700) =2.57). Therefore, based on the following calculations, the maximum AEL of a single pulse is approximately equal to 200 nJ (i.e., 40 watts of peak power):
[0078] AEL 每脉冲 =7.7×10 -8 ×2.57≈200nJ
[0079] As described above, according to the embodiment, the two pulse power levels (i.e., non-uniform pulse power) of the emitted optical pulse are subject to the following conditions:
[0080] a. AEL of non-uniform pulse power 总能量 AEL equal to uniform pulse power 总能量 (i.e. AEL) 总能量 constant);
[0081] b. Maximum pulse power is limited by AEL during pulse time. 每脉冲 .
[0082] use Figures 5B to 5D One of the binary power schemes can determine the pulse power level of an optical pulse as follows, while simultaneously satisfying the two constraints mentioned above:
[0083] L1≈200nJ (i.e., 1.5 times the uniform power level)
[0084] L2≈67nJ (i.e., 0.5 times the uniform power level)
[0085] Assume the number of pixels with power level L1 is the same as the number of pixels with power level L2. Under this assumption, power level L2 can be determined according to the first condition mentioned above (i.e., AEL of non-uniform pulse power). 总能量 AEL equal to uniform pulse power 总能量 As determined above.
[0086] It should be noted that, depending on the power scheme used in the optical pulse detector (e.g., LiDAR), there can be more than two pulse power levels. For example, if a ternary power scheme is used, there are three pulse power levels that satisfy the above constraints.
[0087] The embodiments of the present invention can improve the detection probability of distant objects within the field of view (FOV). Figure 6 The object detection probabilities evaluated for uniform pulse power schemes and non-uniform pulse power schemes under different SNRs are shown. Figure 6In this method, the detection probability of a non-uniform pulse power scheme is calculated as the average of the detection probabilities of the upper and lower limits of the pulse power level.
[0088] refer to Figure 6 When the SNR value is within the high SNR region of 630, the detection probability of both the uniform pulse power scheme and the non-uniform pulse power scheme converges to 1. This means that pulses returning from or backscattering from short-range targets will not be lost. Therefore, the point cloud density of short-range targets will be sufficiently high.
[0089] When the SNR value is within the middle SNR region 620, the detection probability of the non-uniform pulse power scheme is lower than that of the uniform pulse power scheme (i.e., the prior art). However, when the SNR value is within the low SNR region 610, the detection probability of the non-uniform pulse power scheme is higher than that of the uniform pulse power scheme.
[0090] While non-uniform pulse power schemes can achieve high detection probabilities in low SNR regions at the cost of low detection probabilities in intermediate SNR regions, this is advantageous, especially when detecting distant objects. When using a non-uniform pulse power scheme, the detection probability and point cloud density of light pulses reflected or backscattered from distant objects (i.e., objects corresponding to low SNR) are higher than those using a uniform pulse power scheme.
[0091] According to the embodiment, different non-uniform pulse power schemes can be used depending on the importance of objects within the scanned scene or FOV. Furthermore, the pulse power level of the optical pulse can be adjusted according to the importance of objects within the scanned scene or FOV. Adjusting the pulse power level is particularly beneficial for scanned scenes with SNR ranging from the low SNR region 610 to the intermediate SNR region 620.
[0092] Figure 7 Point clouds generated for distant objects using conventional uniform pulse power schemes and non-uniform pulse power schemes are shown. Figure 7 This illustrates how a non-uniform pulsed power scheme can be used to increase the point cloud density of a distant object (e.g., train 400). Scenes 710 and 715 show point clouds generated for train 400 using a conventional uniform pulsed power scheme. Scenes 720 and 725 show point clouds generated for train 400 using a non-uniform pulsed power scheme.
[0093] Figure 7It is clearly shown that in scenes 720 and 725, the point cloud density of train 400 is significantly higher, while in scenes 710 and 715, the point cloud density of street 450 is higher. Considering that train 400 is the target object and is more important than street 450, a non-uniform pulse power scheme can be preferred when scanning the field of view (FOV). Due to AEL limitations, using a conventional uniform pulse power scheme may not increase the point cloud density of distant objects (such as train 400).
[0094] Figure 8 Numerical analysis was performed to compare point cloud densities. Figure 8 The charts in the figure provide numerical analysis of point cloud density at different distances for traditional uniform pulse power schemes and non-uniform pulse power schemes.
[0095] Increasing the detection probability of distant objects means expanding the maximum detection range of objects. For example... Figure 9 As shown, using a non-uniform pulse power scheme with pulse power levels of 1.7 and 0.3 (i.e., 1.7 times the uniform power level and 0.3 times the uniform power level), when the detection probability (P... D When the SNR is 0.2, it will increase by 1 dB. This is because the received power is inversely proportional to the square of the distance (i.e., ...). Increasing the SNR by 1 dB increases the maximum detectable object distance. In simulations, increasing the SNR by 1 dB expanded the maximum detectable object distance range by 12%. Increasing the SNR to a lower level (e.g., 10 dB) shows that the average detection probability of the non-uniform power scheme is greater than that of the uniform pulse power scheme.
[0096] (P D(@功率上限电平) +P D(@功率下限电平) ) / 2>P D(@均匀功率)
[0097] According to the embodiments, as described above, the non-uniform pulse power scheme can be used to increase the detection probability of light pulses returning or backscattered from distant objects. Increasing the detection probability of returning light pulses results in a higher point cloud density for distant objects. Furthermore, the maximum range for detecting objects within the field of view (FOV) is increased. When scanning the field of view (FOV), the pulse power of the non-uniform pulse power scheme must meet eye safety standards, namely, the AEL per pulse (for a single pulse) and the total AEL (for a series of pulses).
[0098] According to embodiments, non-uniform pulse power schemes can improve the detection probability of objects under other adverse conditions. For example, embodiments of the present invention improve the detection probability of objects with low reflectivity or in adverse weather conditions (e.g., foggy and rainy days associated with high atmospheric extinction coefficients). The pulse power of the non-uniform power scheme can be adjusted according to the scene or location of the important object. Various embodiments of the present invention can be implemented using two or more optical pulse generators, which can dynamically adjust the optical power level individually or collaboratively.
[0099] Figure 10 This is a schematic diagram of an electronic device 1000 for detecting objects within a field of view (FOV) according to different embodiments of the present invention. The electronic device 1000 can perform any or all of the methods and features described explicitly or implicitly herein. For example, a time-of-flight (TOF) LiDAR device may be configured with device 1000.
[0100] As shown in the figure, the device includes a processor 1010, a memory 1020, a non-transient mass storage 1030, an I / O interface 1040, a network interface 1050, an optical pulse transmitter 1060, an optical pulse generator 1070, and an optical pulse receiver 1080, all of which are communicatively coupled via a bidirectional bus 1090. According to some embodiments, any or all of the components shown may be used, or only a subset of the components may be used. In some embodiments, the device 1000 may include one or more transceivers for performing the operation of one or more optical pulse transmitters 1060 and one or more optical pulse receivers 1080, rather than including a single one or more optical pulse transmitters and one or more optical pulse receivers. Furthermore, the device 1000 may include multiple instances of some components, such as multiple processors, multiple memories, or multiple transceivers. Additionally, components in the hardware device may be directly coupled to other components without a bidirectional bus.
[0101] Memory 1020 may include any type of non-transitory memory, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), read-only memory (ROM), or any combination thereof. Mass storage 1030 may include any type of non-transitory storage device, such as a solid-state drive, hard disk drive, disk drive, optical disk drive, USB flash drive, or any computer program product for storing data and machine-executable program code. According to some embodiments, memory 1020 or mass storage 1030 may record statements and instructions executable by processor 1010 for performing any of the above-described methods.
[0102] One or more light pulse emitters 1060 can emit multiple light pulses having at least two pulse power levels. For example, the multiple light pulses include a first light pulse with a first pulse power and a second light pulse with a second pulse power. The first pulse power differs from the second pulse power. The power per pulse of the multiple light pulses is less than or equal to a predefined limit, such as the average efficiency per pulse (AEL) (e.g., the average of the total AEL). According to some embodiments, the multiple light pulses may be light pulses generated by a light pulse generator 1070 or one or more external light pulse generators. In various embodiments, the operation of one or more light pulse emitters 1060 can be performed according to statements and instructions executed by a processor 1010. For example, the pulse power level of the light pulses emitted by the emitter 1060 can be controlled by statements and instructions executed by the processor 1010. Additionally, the exposure angle of the light pulse to be emitted (e.g., the exposure angle α240 described above) can be controlled by statements and instructions executed by the processor 1010. In some embodiments, the statements and instructions executed by the processor 1010 can be received from an external controller via a network interface 1050.
[0103] According to some embodiments, the operation of the optical pulse generator 1070 can be executed according to statements and instructions executed by the processor 1010. For example, multiple optical pulses can be generated by the optical pulse generator 1070 according to statements and instructions executed by the processor 1010. In some cases, the pulse power level of the optical pulses can be determined by these statements and instructions. In some embodiments, the statements and instructions executed by the processor 1010 can be received from an external controller via the network interface 1050.
[0104] One or more optical pulse receivers 1080 can receive multiple returned optical pulses. The multiple returned optical pulses are light pulses reflected or backscattered from objects within the FOV. The multiple returned optical pulses can represent information related to the objects within the FOV. The information related to the objects within the FOV can be generated by the processor 1010 from the returned optical pulses. The object-related information may include three-dimensional information, such as a point cloud representing the objects within the FOV. In some embodiments, the generated information can be transmitted to other external devices via a network interface 1050.
[0105] It should be understood that although specific embodiments of the technology have been described herein for illustrative purposes, various modifications may be made without departing from the scope of the technology. The specification and drawings are to be considered merely as an illustration of the invention as defined in the appended claims, and any and all modifications, variations, combinations, or equivalents falling within the scope of this specification are contemplated. Specifically, computer program products or program elements for storing machine-readable signals, or program storage or storage devices such as magnetic or optical, magnetic tape, or optical discs, are provided within the scope of this technology for controlling the operation of a computer according to the methods of this technology and / or constructing some or all of its components according to the systems of this technology.
[0106] The actions associated with the methods described herein can be implemented as coded instructions in a single computer program product. In other words, the computer program product is a computer-readable medium in which software code is recorded to execute the method when the computer program product is loaded into memory and executed on the microprocessor of a wireless communication device.
[0107] The actions associated with the methods described herein can be implemented as coded instructions in multiple computer program products. For example, a first part of the method can be executed by a computing device, and a second part of the method can be executed by other computing devices, servers, etc. In this case, each computer program product is a computer-readable medium on which software code is recorded to execute the appropriate part of the method when the computer program product is loaded into memory and executed on the microprocessor of the computing device.
[0108] Furthermore, each operation of the method can be executed on any computing device (e.g., personal computer, server, PDA, etc.) and is performed based on one or more program elements, modules, or objects generated from any programming language (e.g., C++, Java, etc.), or a portion thereof. Additionally, each operation, or the file or object implementing each operation, can be executed by dedicated hardware or a circuit module designed for this purpose.
[0109] Obviously, the above-described embodiments of the present invention are exemplary and can be implemented in many ways. Such present or future changes should not be considered as a departure from the spirit and scope of the invention, and all such modifications that are obvious to those skilled in the art are intended to be included within the scope of the following claims.
Claims
1. A method for detecting objects within a field of view (FOV), characterized in that, The method includes: A first set of light pulses is emitted, wherein the first set of light pulses includes at least one first light pulse having a first pulse power and at least one second light pulse having a second pulse power, the first pulse power and the second pulse power being determined according to a non-uniform power scheme, the first pulse power being different from the second pulse power, each first light pulse corresponding to a first pixel unit in a pulse power distribution pattern, and each second light pulse corresponding to a second pixel unit in a pulse power distribution pattern; the power per pulse of the first set of light pulses is less than or equal to a predefined limit. Receive a first set of returned light pulses, wherein the first set of returned light pulses represents information related to the object within the FOV; The information relating to the object within the FOV is generated from the first set of returned light pulses.
2. The method according to claim 1, characterized in that, The power of the first pulse is greater than the predefined limit.
3. The method according to claim 1 or 2, characterized in that, The method further includes: The second set of light pulses is emitted, wherein the second set of light pulses includes a third light pulse with a third pulse power and a fourth light pulse with a fourth pulse power, and the power of each pulse of the second set of light pulses is less than or equal to the predefined limit value; Receive a second set of returned light pulses, wherein the second set of returned light pulses represents information related to the object within the FOV; The information relating to the object within the FOV is generated from the second set of returned light pulses.
4. The method according to claim 1 or 2, characterized in that, The method further includes generating the first set of light pulses.
5. The method according to claim 3, characterized in that, The method further includes generating the second set of light pulses.
6. The method according to claim 1 or 2, characterized in that, The first set of light pulses is a laser pulse.
7. The method according to claim 3, characterized in that, The second set of light pulses is a laser pulse.
8. The method according to claim 1 or 2, characterized in that, The information generated in relation to the object within the FOV includes the three-dimensional information of the object within the FOV.
9. The method according to claim 8, characterized in that, The three-dimensional information includes point clouds representing the objects within the FOV.
10. The method according to claim 1 or 2, characterized in that, The predefined limit is the achievable emission limit (AEL) per pulse.
11. The method according to claim 1 or 2, characterized in that, One or more of the first pulse power and the second pulse power are adjustable.
12. The method according to claim 3, characterized in that, One or more of the third pulse power and the fourth pulse power are adjustable.
13. A device for detecting objects within a field of view (FOV), characterized in that, The device includes: One or more optical pulse emitters are used to emit a plurality of optical pulses, wherein the plurality of optical pulses include at least one first optical pulse having a first pulse power and at least one second optical pulse having a second pulse power, the first pulse power and the second pulse power being determined according to a non-uniform power scheme, the first pulse power being different from the second pulse power, each first optical pulse corresponding to a first pixel unit in a pulse power distribution pattern, and each second optical pulse corresponding to a second pixel unit in a pulse power distribution pattern; the power per pulse of the optical pulses is less than or equal to a predefined limit. One or more light pulse receivers are configured to receive a plurality of returned light pulses, wherein the plurality of returned light pulses represent information relating to the object within the FOV; A processor is configured to generate, from the plurality of returned light pulses, the information relating to the object within the FOV.
14. The device according to claim 13, characterized in that, The device is a time-of-flight (TOF) optical detection and ranging LiDAR device.
15. The device according to claim 13 or 14, characterized in that, The device further includes one or more light pulse generators for generating the light pulses.
16. The device according to claim 13 or 14, characterized in that, The light pulse is a laser pulse.
17. The device according to claim 13 or 14, characterized in that, The information generated in relation to the object within the FOV includes the three-dimensional information of the object within the FOV.
18. The device according to claim 17, characterized in that, The three-dimensional information includes point clouds representing the objects within the FOV.
19. The device according to claim 13 or 14, characterized in that, The predefined limit is the achievable emission limit (AEL) per pulse.
20. The device according to claim 13 or 14, characterized in that, One or more of the first pulse power and the second pulse power are adjustable.