EP4469817A1

EP4469817A1 - Lidar apparatus and method of determining height of object with lidar sensor

Info

Publication number: EP4469817A1
Application number: EP22725185.7A
Authority: EP
Inventors: Aron JERRHAGEN; Maxime DEROME; Zhongxia Simon He; Hossein MASHAD NEMATI; Mikhail Ivanov
Original assignee: Huawei Technologies Co Ltd
Current assignee: Shenzhen Yinwang Intelligenttechnologies Co Ltd
Priority date: 2022-04-22
Filing date: 2022-04-22
Publication date: 2024-12-04
Also published as: CN118901024A; WO2023202779A1

Abstract

A lidar apparatus includes a vibration means configured to apply vibrations to a laser part of a lidar sensor to provide sensor pitch variations of an amplitude modulo less than or equal to a resolution of the lidar sensor. The lidar apparatus is used to measure the height of small objects protruding above the ground without using high sensor resolution. The lidar apparatus provides reliable detection of the small objects and accurate height estimation for the small objects at a large distance with reduced cost.

Description

LIDAR APPARATUS AND METHOD OF DETERMINING HEIGHT OF OBJECT WITH LIDAR SENSOR

TECHNICAL FIELD

The present disclosure relates generally to the field of laser systems and more specifically, to a lidar apparatus and a method of determining height of an object with a lidar sensor.

BACKGROUND

In recent years, laser technology has gained huge popularity in various applications such as light detection and ranging (LIDAR or simply lidar) systems and the like. The LIDAR systems provide depth information about a scene based on the time taken by a laser in transmission and receiving to a receiver (or a sensor) after striking a target. The LIDAR systems are generally used in automobile technologies and are considered to be a key sensor for different vehicles, such as for advanced driver-assistance systems (ADAS) and autonomous drive (AD), because of very high resolution and accuracy as compared to other sensor technologies. With the advancement of automotive functions, the demand for the resolution is increasing because detection of small objects on the ground at high speed requires high resolutions. The detection of a small object at high speed is desired to enhance autonomous emergency braking (AEB) systems because a minimum sized small object is also dangerous at high speed. Furthermore, detecting such an object at a long distance (e.g., 100-200 meters) needs time to stop the vehicle at high speed and a certain angular resolution for the detection of the small object. However, with conventional LIDAR systems, it is very challenging and expensive to achieve high resolution when the vehicle is at high speed.

Currently, certain attempts have been made to improve the performance of the conventional LIDAR systems, such as in one attempt, the issue of detecting small objects on the road was partially resolved, but it requires a higher sensor resolution in order to detect the small object at a long-range. Further, ground removal at long ranges is unreliable and requires computational complexity. In general, conventional systems do not provide reliable detection of small objects and are also very expensive. Furthermore, such attempts require at least two points on the small object for reliable detection and separation from the ground. Moreover, such attempts do not provide sufficient height estimation accuracy. As a result, there exists a technical problem of how to measure the height of the small objects protruding above the ground without using high sensor resolution.

Therefore, in light of the foregoing discussion, there exists a need to overcome the aforementioned drawbacks associated with the conventional object detection systems.

SUMMARY

The present disclosure provides a lidar apparatus and a method of determining the height of an object with a lidar sensor. The present disclosure provides a solution to the existing problem of how to measure the height of the small objects protruding above the ground without using high sensor resolution. An objective of the present disclosure is to provide a solution that overcomes at least partially the problems encountered in the prior art and provides an improved lidar apparatus and an improved method for determining the height of an object with a lidar sensor for small object detection, such as to enable autonomous emergency braking (AEB) for small obstacles.

One or more objectives of the present disclosure is achieved by the solutions provided in the enclosed independent claims. Advantageous implementations of the present disclosure are further defined in the dependent claims.

In one aspect, the present disclosure provides a lidar apparatus including a vibration means configured for applying vibrations to a laser part of a lidar sensor to provide sensor pitch variations of an amplitude modulo less than or equal to a resolution of the lidar sensor.

The lidar apparatus uses the vibration means and provides reliable detection of small objects and accurate height estimation for the small objects at a large distance. Further, the lidar apparatus provides reliable ground suppression and low sensitivity to ground imperfections. Further, the sensor vibrations are random, so angles of the sensor pitch vibrations need not to be carefully estimated. In addition, the lidar apparatus detects the small objects from long distances at a low cost and with low resolution.

In an implementation form, the vibration means is configured for applying vibrations to the laser part by applying vibrations to the lidar sensor as a whole. It is advantageous to apply vibration to the whole lidar sensor so that the vibration reaches the laser part without providing vibration separately to the laser part.

In another implementation form, the vibration means is configured for applying vibrations to the transmitting lens of the laser part.

The vibration means is connected to the transmitting lens to provide vibration in the laser that further provides simple hardware design, and the connection is not complex, which provides hassle-free manufacturing. In a further implementation form, the vibration means is configured for applying vibrations to an output mirror of the laser part.

It is advantageous to apply vibration to the output mirror of the laser part that provides vibration in the laser that further provides simple hardware design, and the connection is not complex, which provides hassle-free manufacturing.

In a further implementation form, the sensor pitch vibrations are random and uniformly distributed on an interval from zero to the resolution of the lidar sensor.

The random pitch vibrations of the sensor and uniform distribution of the amplitude modulo on an interval from zero to the resolution of the lidar sensor reduces the cost of the lidar apparatus because a low-resolution lidar is required for the measurement of the small objects.

In another aspect, the present disclosure provides a method of determining a height of an object with a lidar sensor, the method includes applying vibrations to a laser part of a lidar sensor to provide sensor pitch variations of an amplitude modulo less than or equal to a resolution of the lidar sensor. Further, the method includes, obtaining a heat map of a field of view, FOV, based on a point cloud from the lidar sensor, each point in the point cloud corresponds to a detection by the lidar sensor. Further, detecting an object by means of determining an area with a high concentration of points in the heat map. The method further includes, determining a number of detected points and a distance range to the detected object based on the heat map. Further, the method includes, estimating a probability of obtaining the number of detected points from the detected object based on the distance range and a resolution of the lidar sensor. Further, determining a height of the detected object as a difference between the number of detected points and the estimated probability plus one multiplied by the distance range and the resolution. The method discloses determining the height of the object with the lidar sensor. Further, the method discloses applying vibrations to the laser part of the lidar sensor to provide sensor pitch variations of the amplitude modulo less than or equal to a resolution of the lidar sensor that helps to collect statistics values to accurately estimate the height of the object. Further, the heat map for the FOV is obtained by the point cloud from the lidar sensor that helps to detect the object easily by showing a high concentration of the heat. Moreover, the high concentration of points in the heat map detects the object in the area. The concentration of points provides an indication of the object in the heat map. The number of detected points and the distance range to the detected object is determined that is based on the heat map, which helps to identify the object. The distance range and the resolution of the lidar sensor are used to estimate the probability of obtaining the number of detected points from the detected object, which helps in the evaluation of the object. Further, the difference between the number of detected points and the estimated probability plus one multiplied by the distance range and the resolution is obtained to determine the height of the object in a cost- effective manner with a low-resolution lidar. The method further provides accurate measurement of the height of the object that helps to enable autonomous emergency braking (AEB) for small obstacles in a vehicle. Further, the method is not affected by the variation in the pitch of the vehicle. The method also provides low computational complexity with reliable object detection on the ground due to ground detection spread caused by the vibrations. Further, the method does not need an accurate alignment for vertical sensors.

It has to be noted that all devices, elements, circuitry, units, and means described in the present application could be implemented in the hardware elements or any kind of combination thereof. All steps which are performed by the various entities described in the present application as well as the functionalities described to be performed by the various entities are intended to mean that the respective entity is adapted to or configured to perform the respective steps and functionalities. Even if, in the following description of specific embodiments, a specific functionality or step to be performed by external entities is not reflected in the description of a specific detailed element of that entity, which performs that specific step or functionality, it should be clear for a skilled person that these methods and functionalities can be implemented in respective software or hardware elements, or any kind of combination thereof. It will be appreciated that features of the present disclosure are susceptible to being combined in various combinations without departing from the scope of the present disclosure as defined by the appended claims.

Additional aspects, advantages, features, and objects of the present disclosure would be made apparent from the drawings and the detailed description of the illustrative implementations construed in conjunction with the appended claims that follow.

BRIEF DESCRIPTION OF THE DRAWINGS

The summary above, as well as the following detailed description of illustrative embodiments, is better understood when read in conjunction with the appended drawings. For the purpose of illustrating the present disclosure, exemplary constructions of the disclosure are shown in the drawings. However, the present disclosure is not limited to specific methods and instrumentalities disclosed herein. Moreover, those in the art will understand that the drawings are not to scale. Wherever possible, like elements have been indicated by identical numbers.

Embodiments of the present disclosure will now be described, by way of example only, with reference to the following diagrams wherein:

FIG. 1 is a block diagram that depicts a lidar apparatus, in accordance with an embodiment of the present disclosure;

FIG. 2A is another block diagram that depicts a lidar apparatus, in accordance with another embodiment of the present disclosure;

FIG. 2B is another block diagram that depicts a lidar apparatus, in accordance with yet another embodiment of the present disclosure;

FIG. 3 is an illustration that depicts a field of view of a lidar apparatus, in accordance with another embodiment of the present disclosure;

FIG. 4A is a graphical representation that illustrates an example of a vibration signal, in accordance with an embodiment of the present disclosure;

FIG. 4B is a graphical representation that illustrates an example of an estimated pitch distribution, in accordance with an embodiment of the present disclosure;

FIG. 5 A is an illustration that depicts a graphical representation of a distribution of the pitch, in accordance with an embodiment of the present disclosure; FIG. 5B is a graphical representation that illustrates an example of controlled pitch variation, in accordance with an embodiment of the present disclosure;

FIG. 6A is a graphical representation that illustrates a number of lidar detections, in accordance with an embodiment of the present disclosure;

FIG. 6B is a graphical representation that illustrates an accuracy for objects of different sizes, in accordance with an embodiment of the present disclosure;

FIG. 7 depicts a flowchart of height estimation calculation-based algorithm by a lidar apparatus, in accordance with an embodiment of the present disclosure;

FIG. 8 is a graphical representation that illustrates an example of a heat map generated after point cloud accumulation, in accordance with an embodiment of the present disclosure;

FIG. 9 is a graphical representation that illustrates an example of a performance of a height estimation calculation-based algorithm based on the accuracy of the probability estimation, in accordance with an embodiment of the present disclosure; and

FIG. 10 is a method of determining a height of an object with a lidar sensor, in accordance with an embodiment of the present disclosure.

In the accompanying drawings, an underlined number is employed to represent an item over which the underlined number is positioned or an item to which the underlined number is adjacent. A non-underlined number relates to an item identified by a line linking the non-underlined number to the item. When a number is non-underlined and accompanied by an associated arrow, the non-underlined number is used to identify a general item at which the arrow is pointing.

DETAILED DESCRIPTION OF EMBODIMENTS

The following detailed description illustrates embodiments of the present disclosure and ways in which they can be implemented. Although some modes of carrying out the present disclosure have been disclosed, those skilled in the art would recognize that other embodiments for carrying out or practicing the present disclosure are also possible.

FIG. 1 is a block diagram that depicts a lidar apparatus, in accordance with an embodiment of the present disclosure. With reference to FIG. 1, there is shown a block diagram 100 that depicts a light detection and ranging (LIDAR or lidar) apparatus 102. The lidar apparatus 102 includes a vibration means 104 and a lidar sensor 106. The lidar sensor 106 further includes a laser part 108. The laser part 108 further includes a transmitting lens 110 and an output mirror 112.

The lidar apparatus 102 is utilized to measure the height of small objects by analysing a laser beam with vibrations that reflects after striking an object. The lidar apparatus 102 provides depth information of the object. In an implementation, the lidar apparatus 102 is used in the vehicle for assisting the driver to apply brakes upon detecting an obstacle in front of the vehicle. In another implementation, the lidar apparatus 102 is used in the aeroplane to detect the distance from the ground while landing.

The vibration means 104 is used in the lidar apparatus 102 to generate vibration that is further applied on the laser part 108 of the lidar sensor 106. In an example, the vibration means 104 corresponds to piezo actuators and the like.

The lidar sensor 106 is used in the lidar apparatus 102 to measure a physical value from the laser that is received after reflection from the object and turn the measured physical value into an analogue electrical signal. The lidar sensor 106 may include suitable logic, circuitry, and interfaces that are configured to sense the reflected laser. Examples of implementation of the lidar sensor 106 may include but are not limited to a laser distance sensor, a laser photoelectric sensor, a laser edge detection sensor and the like.

The laser part 108 is used to generate the laser for the detection of small objects. In an implementation, the laser part 108 includes the transmitting lens 110 and the output mirror 112. The transmitting lens 110 is used for the transmission of the laser-generated by the laser part 108.

The output mirror 112 is also used in the laser part 108 to amplify the light transmitted by the transmitting lens 110. Examples of implementation of the output mirror 112 may include but are not limited to a spherically curved mirror, a concave mirror, a convex mirror, and the like.

There is provided the lidar apparatus 102 that includes the vibration means 104 configured to apply vibrations to the laser part 108 of the lidar sensor 106 and to provide sensor pitch variations of an amplitude modulo less than or equal to a resolution of the lidar sensor 106. In other words, the vibrations generated by the vibration means 104 are firstly applied to the laser part 108 of the lidar sensor 106. The vibration performs vertical scanning for the lidar apparatus 102, and beneficially as compared to the conventional approach, the vibrations do not need to be controlled precisely. Further, the vibration applied on the laser part 108 provides sensor pitch variations of the amplitude modulo. In an implementation, the amplitude modulo of the sensor pitch variations is less than the resolution of the lidar sensor 106. In an implementation, the amplitude modulo of the sensor pitch variations is equal to the resolution of the lidar sensor 106. Moreover, the resolution of the lidar sensor 106 is selected to keep the probability of receiving one detection on the smallest object is sufficiently high. The sensor pitch variations are beneficial to accurately estimate the height of the object after performing multiple observations.

In an implementation, the lidar apparatus 102 is used as a stand-alone device with a specific application of detecting the small objects from large distances. In another implementation, the lidar apparatus 102 is integrated into a conventional wide field of view (FOV) lidar for scanning mode so that the vibration means 104 also provides sensor stabilization in different modes.

In accordance with an embodiment, the vibration means 104 is configured for applying vibrations to the laser part 108 by applying vibrations to the lidar sensor 106 as a whole. The vibrations are applied on the lidar sensor 106 so that the vibration reaches the laser part 108. In an implementation, the vibration is provided on the lidar sensor 106 to decrease the complexity of connecting the vibration means 104 to the lidar apparatus 102 and improve hardware design.

In accordance with an embodiment, the sensor pitch vibrations are random, and the amplitude modulo are uniformly distributed on an interval from zero to the resolution of the lidar sensor 106. The amplitude modulo is uniformly distributed, and the sensor pitch vibrations are random, so the angles of the sensor pitch vibrations do not need to be carefully estimated. Further, the frequency of the vibration is to be significantly larger than the frame rate divided by the number of measurements. For example, the frequency required to be larger than 5 hertz (Hz) in a lidar having a frame rate of 200 Hz and a number of measurements of 40 to detect an object of 30 cm at 100 m. Moreover, frequencies up to 100 Hz are sufficient. Further, an amplitude of the vibration requires to be larger than the resolution divided by two but kept as small as possible to ensure that the lidar is pointing to the region of interest. In an implementation, the required vibration may be achieved by using piezo actuators as they are cheap, reliable, and provides sufficient motion frequency and range.

The lidar apparatus 102 provides reliable detection of the small objects with accurate height estimation for the small objects at a large distance. Further, the lidar apparatus 102 uses the vibration means 104 that provides reliable ground suppression and low sensitivity to ground imperfections. Further, the sensor vibrations are random, so angles of the sensor pitch vibrations need not be carefully estimated. In addition, the lidar apparatus 102 is capable to detect the small objects from long distances with reduced cost by using lidar of low resolution.

FIG. 2A is another block diagram that depicts a lidar apparatus, in accordance with another embodiment of the present disclosure. FIG. 2A is shown in conjunction with elements from FIG.1. With reference to FIG. 2A, there is shown a block diagram 200A that depicts the lidar apparatus 102. The lidar apparatus 102 includes the vibration means 104, the lidar sensor 106, the laser part 108, the transmitting lens 110, the output mirror 112, and a receiving lens 202. The receiving lens 202 is used to receive the light that is reflected after striking the object. Examples of the receiving lens 202 may include but are not limited to a spherically curved lens, a concave lens, a convex lens, and the like.

In accordance with an embodiment, the vibration means 104 is configured to apply vibrations to the transmitting lens 110 of the laser part 108. Firstly, the laser part 108 of the lidar apparatus 102 generates the laser for the detection of the small object. Further, the laser generated by the laser part 108 is transmitted by the transmitting lens 110. Moreover, the transmitting lens 110 keeps the laser in a straight line by forming a beam that protects the laser from being scattered. In addition, the vibration means 104 is connected to the transmitting lens 110 to provide vibration in the laser that provides simple hardware design, and the connection is not complex, which provides hassle-free manufacturing. Further, the laser with the vibrations is projected out through the output mirror 112 to strike the object. After striking the object, the laser with the vibrations reflects and is received by the receiving lens 202. Furthermore, the receiving lens provides the received reflected laser to the lidar sensor 106 for the detection of the object. FIG. 2B is another block diagram that depicts a lidar apparatus, in accordance with yet another embodiment of the present disclosure. FIG. 2B is shown in conjunction with elements from FIG. 1 and FIG. 2A. With reference to FIG. 2B, there is shown a block diagram 200B that depicts the lidar apparatus 102. The lidar apparatus 102 includes the vibration means 104, the lidar sensor 106, the laser part 108, the transmitting lens 110, the output mirror 112, and the receiving lens 202.

In accordance with an embodiment, the vibration means 104 is configured to apply vibrations to the output mirror 112 of the laser part 108. Firstly, the laser part 108 of the lidar apparatus 102 generates the laser for the detection of the small object. Further, the laser generated by the laser part 108 is transmitted by the transmitting lens 110. Moreover, the transmitting lens 110 keeps the laser in a straight line by forming a beam that protects the laser from being scattered. After that, the output of the transmitting lens 110 is received by the output mirror 112. Further, the vibration means 104 is connected to the output mirror 112 to provide vibration in the laser that provides simple hardware design, and the connection is not complex, which provides hassle-free manufacturing. Thereafter, the laser with the vibrations is projected out through the output mirror 112 to strike the object. After striking the object, the laser with the vibrations reflects and is received by the receiving lens 202. Further, the receiving lens provides the received reflected laser to the lidar sensor 106 for the detection of the object.

FIG. 3 is an illustration that depicts a field of view of a lidar apparatus, in accordance with an embodiment of the present disclosure. FIG. 3 is shown in conjunction with elements from FIG. 1, FIG. 2A and FIG. 2B. With reference to FIG. 3, there is shown an illustration 300 that depicts the lidar apparatus 102. The lidar apparatus 102 is used to detect an object 302. The object 302 may be any small obstacle that may include but is not limited to stone, pebbles, potholes and alike.

The lidar apparatus 102 is placed at a certain height, due to which the field of view (FOV) of the lidar apparatus 102 is increased. Further, the laser is generated by the lidar apparatus 102 with the vibration to detect the height of the object 302 that is placed in the field of view (FOV) of the lidar apparatus 102. The vibration with the laser emits out of the lidar apparatus 102 helps to collect statistics and accurately estimate the height of the object 302. Further, multiple dots present in front of the object 302 represents points generated due to the striking of the laser with vibrations on the ground. Further, the dot on the object 302 represents the point to detect the height of the object 302. Moreover, the points behind the object 302 represent the area beyond the detection range of the lidar apparatus 102. Further, the vibration in the laser helps to generate the points and, apart from providing the desired point statistics for height estimation, significantly simplifies the problem of detection of the object 302 that is lying on the ground. As the object 302 is detected and a range to the object 302 is obtained, then the points from the object 302 are analysed, and the probability Pr{A} is estimated, as further shown and described in FIG. 6A. Further, the height of the object 302 is estimated as further shown and described by using the equation disclosed in FIG. 7. The equation provides the height estimation when all accumulated measurements are obtained at the same range. Further, the height estimation is performed by the height estimation algorithm. Furthermore, on the bases of the height estimation algorithm, the result is obtained that the necessary number of accumulated frames is in the range of 20 to 40. In addition, the frame rate is required to be in the range of 100-200 hertz (Hz) for automotive requirements like detection delay. Moreover, the FOV is narrow both vertically and horizontally to detect small objects at a long-range. Further, the high frame rate is achieved easily by using very limited FOV.

FIG. 4A is a graphical representation that illustrates an example of a vibration signal, in accordance with an embodiment of the present disclosure. FIG. 4A is described in conjunction with elements from FIGs. 1 to 3. With reference to FIG. 4A, there is shown a graphical representation 400A with an X-axis 402A that represents the time (in seconds) and a Y-axis 402B that represents a pitch (in degrees). In an example, the pitch corresponds to the sensor pitch variations.

In the graphical representation 400A, a line 404 represents the vibration signal generated by the vibration means 104. Further, the graphical representation 400A shows that the vibration signals are random. Moreover, the generated vibration signals are used for accurate height estimation, as further shown in FIG. 7.

FIG. 4B is a graphical representation that illustrates an example of an estimated pitch distribution, in accordance with an embodiment of the present disclosure. FIG. 4B is described in conjunction with elements from FIGs. 1 to 4A. With reference to FIG. 4B, there is shown a graphical representation 400B with an X-axis 402A that represents the pitch (in degrees) and a Y-axis 402B that represents sensor pitch variations through a histogram.

In the graphical representation 400B, there is shown a histogram bar 408 that represents the estimated pitch distribution on the interval from zero (0) to thirty-four degrees. With reference to FIG. 4B there is shown that the resulting distribution is sufficiently uniform, as shown via different histogram bars. In an example, sampling timing errors are used to make the distribution even more uniform. The estimated pitch distribution is accurate for the height estimation. In an example, the frequency is around 200 Hz, and the degree resolution is given as «=0.34. The pitch signal can be calculated using the below equation (1):

9(t) = 0.11 (sin(2nft) + sin(2n7.5ft) + sin(2n9.3ft)) (1) where, “f” =10.1 Hz. In an implementation, piezo actuators are used for desired vibrations. Moreover, the piezo actuators are cheap, reliable, and provide sufficient motion frequency and range.

FIG. 5 A is an illustration that depicts a graphical representation of a distribution of the pitch, in accordance with different embodiments of the present disclosure. FIG. 5 A is shown in conjunction with elements from FIG. 1 to FIG. 4. With reference to FIG. 5 A, there is shown a graphical representation 500A with an X-axis 502A that represents the resolution and a Y- axis 502B that represents the pitch distribution.

The graphical representation 500A of the pitch distribution discloses that the vibration provides random pitch variations so that the amplitude modulo is uniformly distributed on the interval from zero to the resolution. Further, the graphical representation 500A discloses the desired distribution of the pitch of the vibration. In an implementation, the required vibration may be achieved by using piezo actuators as they are cheap, reliable, and provide sufficient motion frequency and range.

FIG. 5B is a graphical representation that illustrates an example of controlled pitch variation, in accordance with an embodiment of the present disclosure. FIG. 5B is described in conjunction with elements from FIGs. 1 to 5 A. With reference to FIG. 5B, there is shown a graphical representation 500B with an X-axis 506A that represents the time (in seconds) and a Y-axis 506B that represents the pitch resolution.

In the graphical representation 500B, a first line 508 (i.e., represented as a dotted line) represents the resolution of the lidar sensor, and a second line 510 represents the sensor pitch variations of the amplitude modulo. Further, the graphical representation 500B represents that the value of the amplitude modulo is between zero (0) and the value of the resolution of the lidar sensor. Therefore, the pitch distribution with the value of the amplitude modulo between zero (0) and the value of the resolution of the lidar sensor improves the performance of the lidar apparatus 102 for height estimation of the object.

FIG. 6A is a graphical representation that illustrates a number of lidar detections, in accordance with an embodiment of the present disclosure. FIG. 6A is described in conjunction with elements from FIGs. 1 to 5B. With reference to FIG. 6A, there is shown a graphical representation 600A with an X-axis 602A that represents the range (in meter) and a Y-axis 602B that represents the lidar points.

In the graphical representation 600A, a first line 604 represents “N” number of lidar points, and a second line 606 represents “N+l” number of lidar points. In an implementation, the lidar apparatus 102 produces the “N” number of lidar points. In another implementation, the lidar apparatus 102 produces the “N+l” number of lidar points. Further, the number of lidar points (i.e., the number of lidar points “N”) depends upon the height and the range of the object. In an implementation, if the range is sufficiently large enough, then, in that case, the number of lidar points (i.e., “N”) is equal to zero (0). Finally, the probability of “N” number of detections for the uniform distributed pitch is calculated by using the equation (2) as shown below: where “A” is the height of the object, “r” is the range of the object, "a" is the sensor resolution, and “[•]” represents the ceiling operation. In an example, if the number of detections is equal to one (1), then, in that case, the Pr{l} = In another example, for the detection of the object of the height of 0.30m with the range of 100m, and the probability of 0.5, and the resolution of 0.34 degree is required. Thus, the probability of the “N” number of detections is calculated to further calculate the height of the object accurately.

FIG. 6B is a graphical representation that illustrates an accuracy for objects of different sizes, in accordance with an embodiment of the present disclosure. FIG. 6B is described in conjunction with elements from FIGs. 1 to 6A. With reference to FIG. 6B, there is shown a graphical representation 600B with an X-axis 608A that represents the range (in meters) and a Y-axis 608B that represents the probability of “N” number of detections. In the graphical representation 600B, a first line 610 represents the probability of “N” number of detections for the number of lidar points detected in the range (as shown in FIG. 6B) of 0 to 120.

FIG. 7 depicts a flowchart of height estimation calculation-based algorithm by a lidar apparatus, in accordance with an embodiment of the present disclosure. FIG. 7 is described in conjunction with elements from FIGs. 1 to 6B. With reference to FIG. 7, there is shown a flowchart 700 that includes a series of operations from 702-to-716. In an implementation, the lidar apparatus 102 (of FIG. 1) is configured to execute the flowchart 700.

With reference to FIG. 7, there is shown that at operation 702, the lidar apparatus 102 starts the height estimation calculation-based algorithm. The lidar apparatus 102 first calculates the probability of the “N” number of detections (i.e., Pr {N}) and further calculates the height (i.e., h) of the object accurately. At operation 704, the lidar apparatus 102 obtains the vehicle’s speed and a yaw rotation to determine a lidar point cloud accumulation. In an example, the yaw rotation corresponds to a movement around a yaw axis of a vehicle. The lidar point cloud accumulation is used to calculate the probability of the “N” number of detections (i.e., Pr{A}). Alternately, the lidar apparatus 102 obtains the lidar point cloud information, such as at operation 706. Further, at operation 708, the lidar apparatus 102 accumulates the point cloud in the x-y plane to obtain the heat map (such as the heat map as shown in FIG. 8). Further, at operation 710, the lidar apparatus 102 performs object detection. The heat map obtained at operation 708 provides areas with a high concentration of detections that corresponds to the objects. The object detection provides the range of objects (i.e., the value of “r”). Further, at operation 712, the lidar apparatus 102 performs the statics estimation that includes the estimation of the probability of “N” number of detections (i.e., Pr{A}). The number of detections (i.e., “N”) varies from zero (0) to one (1). Finally, at operation 714, the lidar apparatus 102 calculates the height (i.e., h) of the object. In an implementation, the lidar apparatus 102 finishes the height estimation calculation at operation 716 by using equation (3) as shown below: h = (IV + 1 - Pr V}) ■ a ■ r (3) where “JV” is the number of detections that ranges from zero (0) to one (1), Pr{N} is the probability of “N” number of detections, "a” is equal to 0.34-degree resolution, and “r” corresponds to a range of objects.

It should be noted that the use of the equation (3) is limited to stationary /low speed scenarios. In high-speed scenarios, the height calculation shall also recognize that ‘r’ is changed from measurement to measurement.

In an example, the probability of the number of detections is estimated as 0.45, and the number of detections is zero (0). Then, in that case, the height estimation is calculated as: h = (1 - Pr{0})ar = (1 - 0.45) ■ 0.3/180 ■ n ■ 82 = 0.26m

FIG. 8 is a graphical representation that illustrates an example of a heat map generated after point cloud accumulation, in accordance with an embodiment of the present disclosure. FIG. 8 is described in conjunction with elements from FIGs. 1 to 7. With reference to FIG. 8, there is shown a graphical representation 800 with an X-axis 802A that represents the range (in meters) to a lidar in the longitudinal direction, wherein the lidar position is at point (0,0), and a Y-axis 802B that represents a distance (in meters) from the lidar in the transverse direction, i.e., to sides from the lidar position.

In the graphical representation 800, different dark spots represent the areas with concentrated lidar points. The object is detected by the accumulation of lidar points due to the accumulation of detections from the object, as shown in a dotted circle 804. In an implementation, forty (40) measurements are accumulated for more accuracy without limiting the scope of the present disclosure for the object detection. Further, the graphical representation 800 represents the detection of the object, represented by the dotted circle 804, at the distance of around eighty -two (82) meters from the lidar., After the detection of the object, the probability of “N” number of detections (i.e., Pr{/V}) is estimated to finally estimate the height of the object with more accuracy.

FIG. 9 is a graphical representation that illustrates an example of a performance of a height estimation calculation-based algorithm for a stationery case based on the accuracy of the probability estimation, in accordance with an embodiment of the present disclosure. FIG. 9 is described in conjunction with elements from FIGs. 1 to 8. With reference to FIG. 9, there is shown a graphical representation 900 with an X-axis 902A that represents the range (in meters) and a Y-axis 902B that represents the root-mean square error (RMSE) of the height estimation (in meters) being a measure of the height estimation performance. The graphical representation 900 is obtained for the number of measurements (sesor frames) of 40.

In the graphical representation 900, a first line 904 represents the height estimation performance with respect to the object with the height of 0.3 meters for the number of measurements of 40. Furthermore, a second line 906 represents the height estimation performance with respect to the object with the height of 0.5 meters for the same number of measurements. Furthermore, a third line 908 represents the height estimation performance with respect to the object with the height of 0.7 meters for the same number of measurements. Thus, the first line 904, the second line 906, and the third line 908 correspond to the accuracy for the estimation of the height of the different objects at different distance ranges. The accuracy of the probability estimation of the height of the object is obtained through the number of detected lidar points from the detected object based on the distance range and depends upon the range of the object. Moreover, the accuracy of the probability estimation of the height increases with the increase in the number of measurements.

FIG. 10 is a method of determining a height of an object with a lidar sensor, in accordance with different embodiments of the present disclosure. FIG. 10 is shown in conjunction with elements from FIG. 1 to FIG. 9. With reference to FIG. 10, there is shown a flow chart of the method 1000 for determining a height of an object with a lidar sensor. The method 1000 includes steps 1002 to 1012.

There is provided the method 1000 of determining the height of the object with the lidar sensor. The method 1000 provides reliable detection of the small objects and accurate height estimation for the small objects at a large distance. The method 1000 discloses measuring the height of the small objects protruding above the ground without using high sensor resolution.

At step 1002, the method 1000 includes applying vibrations to the laser part 108 of the lidar sensor 106 to provide sensor pitch variations of an amplitude modulo less than or equal to a resolution of the lidar sensor 106. In other words, the vibrations generated by the vibration means 104 are firstly applied to the laser part 108 of the lidar sensor 106. The vibration performs vertical scanning for the lidar apparatus 102, and beneficially as compared to the conventional approach, the vibrations do not need to be controlled precisely. Further, the vibration applied on the laser part 108 provides sensor pitch variations of the amplitude modulo. In an implementation, the amplitude modulo of the sensor pitch variations is less than the resolution of the lidar sensor 106. In an implementation, the amplitude modulo of the sensor pitch variations is equal to the resolution of the lidar sensor 106. Moreover, the resolution of the lidar sensor 106 is selected to keep the probability of receiving one detection on the smallest object is sufficiently high. The sensor pitch variations are beneficial to accurately estimating the height of the object after performing multiple observations. In an implementation, the lidar apparatus 102 is used as a stand-alone device with a specific application of detecting the small objects from large distances. In another implementation, the lidar apparatus 102 is integrated into a conventional wide field of view (FOV) lidar for scanning mode so that the vibration means 104 also provides sensor stabilization in different modes.

At step 1004, the method 1000 includes obtaining a heat map of afield of view, FOV, based on a point cloud from the lidar sensor 106, each point in the point cloud corresponds to a detection by the lidar sensor 106. The heat map of the FOV is obtained that is based on the point cloud from the lidar sensor 106. The detection performed by the lidar sensor 106 is represented by the point cloud. The method 1000 first calculates the probability of the “N” number of detections (i.e., Pr{N}) and further calculates the height (i.e., h~) of the object accurately. In an implementation, if the lidar sensor 106 is installed in a vehicle, then the method 1000 obtains the vehicle’s speed and a yaw rotation to determine a lidar point cloud accumulation. The lidar point cloud accumulation is used to calculate the probability of the “N” number of detections (i.e., Pr{N}). Further, the method 1000 accumulates the point cloud in the x-y plane to obtain the heat map. At step 1006, the method 1000 includes detecting an object by means of determining an area with a high concentration of points in the heat map. The heat map provides areas with a high concentration of detections that corresponds to the objects. The object detection provides the range of objects (i.e., the value of “r”). Further, the method 1000 performs the statics estimation that includes the estimation of the probability of “N” number of detections (i.e., Pr{N}). The number of detections (i.e., “N”) varies from zero (0) to one (1).

At step 1008, the method 1000 includes determining a number of detected points and a distance range to the detected object based on the heat map. After obtaining the heat map, the number of detected points is determined, and the distance to the detected object is also determined based on the heat map. In an implementation, the detected points present in front of the small objects represent points generated due to the striking of the laser with vibrations on the ground. In yet another implementation, the detected points on the small object represent the point that is to detect the small object. Moreover, the detected points after the small object represent the area beyond the detection range.

At step 1010, the method 1000 includes estimating a probability of obtaining the number of detected points from the detected object based on the distance range and a resolution of the lidar sensor 106. The distance range from the object and the resolution are used in the estimation of obtaining the number of detected points from the detected object. Further, the probability of the number of detections increases if the value of the number of detected points is equal to one.

At step 1012, the method 1000 includes determining a height of the detected object as a difference between the number of detected points and the estimated probability plus one multiplied by the distance range and the resolution. To determine the height of the detected object, the estimated probability plus one multiplied by the distance range and the resolution is subtracted from the number of detected points. After the detection of the object, the probability of “N” number of detections (i.e., Pr {/V} ) is estimated to finally estimate the height of the object with more accuracy. According to an example, the resolution of the lidar sensor 106 is 0.34 degrees, which is required to detect an object of 0.3 metres (m) height at the range of 100 meters (m). The sensor pitch variations are used to accurately estimate the height after performing multiple observations. In an implementation, the object of a given height of the object that is 0.3m at a given range of 100m, the method 1000 generates N detected points, where N is dependent on the height and the range. In another implementation, the object of a given height of the object that is 0.3m at a given range of 100m, the method 1000 generates N + 1 detected points. Moreover, if the range is sufficiently large, N = 0. Further, an analytical expression for the probability of observing N detections is obtained.

In accordance with an embodiment, the lidar sensor 106 is arranged on a vehicle, and the method further includes obtaining the point cloud from the lidar sensor 106 during a movement of the vehicle. The point clouds are obtained during the movement of the vehicle. It can be noted that during the movement of the vehicle, compensation for the motion and to estimate the statistics. In accordance with an embodiment, the lidar sensor 106 is arranged on the vehicle. The lidar sensor 106 is arranged on the vehicle to obtain the information about the vehicle. The information is further required to calculate the probability of the “N” number of detections (i.e., Pr{N}) and further calculate the height (i.e., h~) of the object accurately. The method 1000 includes obtaining the point cloud by the lidar sensor 106 during a movement of the vehicle. The point cloud is obtained by the lidar sensor 106 to further calculate the height of the object. In addition, the method 1000 provides low sensitivity to the pitch of the vehicle and the lidar sensor 106 needs not to be carefully aligned. Further, the method 1000 helps to enable autonomous emergency braking (AEB) at high speed by detecting small objects in the path of the vehicle.

In accordance with an embodiment, obtaining parameters of the movement of the vehicle including a speed and a yaw, the obtaining of the heat map of the scene based on the point cloud includes compensating the movement of the vehicle based on the parameters. Further, the method 1000 includes obtaining parameters of the movement of the vehicle including the speed and the yaw. The method 1000 obtains the vehicle’s speed and a yaw rotation to determine a lidar point cloud accumulation. In an example, the yaw rotation corresponds to a movement around the yaw axis of a vehicle. The lidar point cloud accumulation is used to calculate the probability of the “N” number of detections (i.e., Pr{N}). Further, the method 1000 includes obtaining the heat map of the scene based on the point cloud includes compensating for the movement of the vehicle based on the parameters. The lidar sensor 106 accumulates the point cloud in the x-y plane to obtain the heat map. In addition, the method 1000 enables L3 functions at high speed, such as highway pilot. In an example, the L3 function helps in decreasing driving stress as the driver does not need to be attentive while driving because the vehicle applies brakes automatically upon detecting any obstacle in the path.

In accordance with an embodiment, applying vibrations to the laser part 108 of the lidar sensor 106 includes applying vibrations to the lidar sensor 106 as a whole. The vibrations are applied on the lidar sensor 106 so that the vibration reaches the laser part 108. In an implementation, the vibration is provided on the lidar sensor 106 to decrease the complexity of connecting the vibration means 104 to the lidar apparatus 102 and improve hardware design.

In accordance with an embodiment, applying vibrations to the laser part 108 of the lidar sensor 106 comprises applying vibrations to transmitting lens 110 of the laser part 108. Firstly, the laser part 108 of the lidar apparatus 102 generates the laser for the detection of the small object. Further, the laser generated by the laser part 108 is transmitted by the transmitting lens 110. Moreover, the transmitting lens 110 keeps the laser in a straight line by forming a beam that protects the laser from being scattered. In addition, the vibration means 104 is connected to the transmitting lens 110 to provide vibration in the laser that provides simple hardware design, and the connection is not complex, which provides hassle- free manufacturing. Further, the laser with the vibrations is projected out through the output mirror 112 to strike the object. After striking the object, the laser with the vibrations reflects and is received by the receiving lens 202. Furthermore, the receiving lens provides the received reflected laser to the lidar sensor 106 for the detection of the object.

In accordance with an embodiment, applying vibrations to the laser part of the lidar sensor comprises applying vibrations to an output mirror 112 of the laser part 108. Firstly, the laser part 108 of the lidar apparatus 102 generates the laser for the detection of the small object. Further, the laser generated by the laser part 108 is transmitted by the transmitting lens 110. Moreover, the transmitting lens 110 keeps the laser in a straight line by forming a beam that protects the laser from being scattered. After that, the output of the transmitting lens 110 is received by the output mirror 112. Further, the vibration means 104 is connected to the output mirror 112 to provide vibration in the laser that provides simple hardware design, and the connection is not complex, which provides hassle-free manufacturing. Thereafter, the laser with the vibrations is projected out through the output mirror 112 to strike the object. After striking the object, the laser with the vibrations reflects and is received by the receiving lens 202. Further, the receiving lens provides the received reflected laser to the lidar sensor 106 for the detection of the object.

In accordance with an embodiment, the sensor pitch vibrations are random, and amplitude modulo is uniformly distributed on an interval from zero to the resolution of the lidar sensor 106. The amplitude modulo is uniformly distributed, and the sensor pitch vibrations are random, so the angles of the sensor pitch vibrations do not need to be carefully estimated. Further, the frequency of the vibration is to be significantly larger than the frame rate divided by the number of measurements. For example, the frequency required to be larger than 5 hertz (Hz) to detect an object of 30 cm at 100 m. Moreover, the frequencies up to 100 Hz are sufficient. Further, an amplitude of the vibration requires to be larger than the resolution divided by two but kept as small as possible to ensure that the lidar is pointing to the region of interest. In an implementation, the required vibration may be achieved by using piezo actuators as they are cheap, reliable, and provides sufficient motion frequency and range.

The method 1000 provides reliable detection of the small objects and accurate height estimation for the small objects at the large distance. Further, the method 1000 discloses using the vibration means 104 to provide reliable ground suppression and low sensitivity to ground imperfections. Further, the sensor vibrations are random, so angles of the sensor pitch vibrations need not be carefully estimated. In addition, the method 1000 provides low computational complexity for object detection and height estimation.

Modifications to embodiments of the present disclosure described in the foregoing are possible without departing from the scope of the present disclosure as defined by the accompanying claims. Expressions such as "including", "comprising", "incorporating", "have", "is" used to describe and claim the present disclosure are intended to be construed in a non-exclusive manner, namely allowing for items, components or elements not explicitly described also to be present. Reference to the singular is also to be construed to relate to the plural. The word "exemplary" is used herein to mean "serving as an example, instance or illustration". Any embodiment described as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments or to exclude the incorporation of features from other embodiments. The word "optionally" is used herein to mean "is provided in some embodiments and not provided in other embodiments". It is appreciated that certain features of the present disclosure, which are, for clarity, described in the context of separate embodiments, may also be provided in combination in a single embodiment. Conversely, various features of the invention, which are, for brevity, described in the context of a single embodiment, may also be provided separately or in any suitable combination or as suitable in any other described embodiment of the disclosure.

Claims

1. A lidar apparatus (102) comprising a vibration means (104) configured for applying vibrations to a laser part (108) of a lidar sensor (106) to provide sensor pitch variations of an amplitude modulo less than or equal to a resolution of the lidar sensor (106).

2. The lidar apparatus (102) according to claim 1, wherein the vibration means (104) is configured for applying vibrations to the laser part (108) by applying vibrations to the lidar sensor (106) as a whole.

3. The lidar apparatus (102) according to claim 1, wherein the vibration means (104) is configured for applying vibrations to transmitting lens (110) of the laser part (108).

4. The lidar apparatus (102) according to claim 1, wherein the vibration means (104) is configured for applying vibrations to an output mirror (112) of the laser part (108).

5. The lidar apparatus (102) according to any of claims 1 to 4, wherein the sensor pitch vibrations are random and the amplitude modulo are uniformly distributed on an interval from zero to the resolution of the lidar sensor (106).

6. A method (1000) of determining a height of an object with a lidar sensor (106), the method comprising: applying vibrations to a laser part (108) of a lidar sensor (106) to provide sensor pitch variations of an amplitude modulo less than or equal to a resolution of the lidar sensor (106), obtaining a heat map of a field of view, FOV, based on a point cloud from the lidar sensor (106), wherein each point in the point cloud corresponds to a detection by the lidar sensor (106), detecting an object by means of determining an area with a high concentration of points in the heat map, determining a number of detected points and a distance range to the detected object based on the heat map, estimating a probability of obtaining the number of detected points from the detected object based on the distance range and a resolution of the lidar sensor (106), and determining a height of the detected object as a difference between the number of detected points and the estimated probability plus one multiplied by the distance range and the resolution.

7. The method (1000) of claim 6, wherein the lidar sensor (106) is arranged on a vehicle and the method further comprises: obtaining the point cloud by from the lidar sensor (106) during a movement of the vehicle, and obtaining parameters of the movement of the vehicle including a speed and a yaw, wherein the obtaining of the heat map of the scene based on the point cloud comprises compensating the movement of the vehicle based on the parameters.

8. The method (1000) of claim 6 or 7, wherein applying vibrations to the laser part (108) of the lidar sensor (106) comprises applying vibrations to the lidar sensor (106) as a whole.

9. The method (1000) of claim 6 or 7, wherein applying vibrations to the laser part (108) of the lidar sensor (106) comprises applying vibrations to transmitting lens (110) of the laser part (108).

10. The method (1000) of claim 6 or 7, wherein applying vibrations to the laser part (108) of the lidar sensor (106) comprises applying vibrations to an output mirror (112) of the laser part (108).

11. The method (1000) of any of claims 6 to 10, wherein the sensor pitch vibrations are random and the amplitude modulo are uniformly distributed on an interval from zero to the resolution of the lidar sensor (106).