METHOD FOR PROCESSING DATA PROVIDED BY A LIDAR AND ASSOCIATED COMPUTER

Abstract

A method for processing data provided by a lidar. Obtaining, for measurement points of a lidar, data representative of a diffuse intensity and a distance between the measurement point and the lidar; determining an angle of incidence of the lidar at each measurement point; calculating, for each measurement point, reflectivity from the angle of incidence, the distance and the diffuse intensity; identifying spatially adjacent measurement points having a reflectivity within a given range of values to form common reflectivity zones; defining a cost function for at least one common reflectivity zone, the cost function including a first diffuse intensity and a second diffuse intensity, the angle and the distance of the measurement point in question; minimizing the cost function in order to update at least one of the angle, the distance and the reflectivity at the measurement point, for each of the measurement points belonging to a common reflectivity zone.

Claims

1. A computer configured for: obtaining, for a plurality of measurement points of a lidar, data representative of a diffuse intensity at a measurement point and of a distance between the measurement point and the lidar; determining the direction normal to the surface tangential to each measurement point, and deducing therefrom an angle of incidence of the lidar at each measurement point; calculating, for each measurement point, the reflectivity from the angle of incidence, the distance and the diffuse intensity; identifying spatially adjacent measurement points having a reflectivity within a given range of values centered on a common reflectivity value, in order to form common reflectivity zones; defining a cost function for at least one common reflectivity zone, the cost function comprising a plurality of terms, each term being relative to a measurement point and comprising a first diffuse intensity obtained using the data representative of the diffuse intensity and a second diffuse intensity determined as a function of the common reflectivity value, the angle and the distance of the measurement point in question; and minimizing the cost function, using an iterative non-linear optimization algorithm, in order to update at least one of the angle, the distance and the reflectivity at the measurement point, for each of the measurement points belonging to a common reflectivity zone.

2. The computer as claimed in claim 1, wherein the measurement points belonging to common reflectivity zones also have angles of incidence within a range of permitted values.

3. The computer as claimed claim 1, wherein the diffuse intensity, the angle and the distance are linked by a diffuse intensity calculation function, and the second diffuse intensity is determined using said function.

4. The computer as claimed in claim 3, wherein the diffuse intensity calculation function is written in the form: $\begin{array}{cc}\mathrm{Id}=\frac{Rd\cos \theta }{d^2}& \left[\mathrm{Math}.2\right]\end{array}$ where Id is the diffuse intensity, Rd is the reflectivity, θ is the angle of incidence and d is the distance.

5. The computer as claimed in claim 1, wherein the cost function is equal to the sum of the different terms relative to a measurement point.

6. The computer as claimed in claim 5, wherein the cost function fc, for a common reflectivity zone Rc in question, is written thus: $\begin{array}{cc}\mathrm{fc}=\underset{k=1}{\overset{N}{.Math.}}.Math.{\mathrm{Id}}_k-\frac{\mathrm{Rc}\cos {\theta }_k}{d_k^2}.Math.& \left[\mathrm{Math}.3\right]\end{array}$ where Id.sub.k is the diffuse intensity, d.sub.k is the distance and θ.sub.k is the angle of incidence for a measurement point in question with the index k belonging to the common reflectivity zone in question, and N is the number of measurement points belonging to the common reflectivity zone in question.

7. A motor vehicle provided with a lidar and a computer as claimed in claim 1.

8. A method for processing data provided by a lidar, the method being implemented by a computer and comprising: obtaining, for a plurality of measurement points of a lidar, data representative of a diffuse intensity at a measurement point and of a distance between the measurement point and the lidar; determining the direction normal to the surface tangential to each measurement point, and deducing therefrom an angle of incidence of the lidar at each measurement point; calculating, for each measurement point, the reflectivity from the angle of incidence, the distance and the diffuse intensity; identifying spatially adjacent measurement points having a reflectivity within a given range of values centered on a common reflectivity value, in order to form common reflectivity zones; defining a cost function for at least one common reflectivity zone, the cost function comprising a plurality of terms, each term being relative to a measurement point and comprising a first diffuse intensity obtained using the data representative of the diffuse intensity and a second diffuse intensity determined as a function of the common reflectivity value, the angle and the distance of the measurement point in question; and minimizing the cost function, using an iterative non-linear optimization algorithm, in order to update at least one of the angle, the distance and the reflectivity at the measurement point, for each of the measurement points belonging to a common reflectivity zone.

9. The method as claimed in claim 8, wherein the measurement points belonging to common reflectivity zones also have angles of incidence within a range of permitted values.

10. The method as claimed in claim 8, wherein the diffuse intensity, the angle and the distance are linked by a diffuse intensity calculation function, and the second diffuse intensity is determined using said function.

11. The method as claimed in claim 8, wherein the diffuse intensity calculation function is written in the form: $\begin{array}{cc}\mathrm{Id}=\frac{Rd\cos \theta }{d^2}& \left[\mathrm{Math}.2\right]\end{array}$ where Id is the diffuse intensity, Rd is the reflectivity, θ is the angle of incidence and d is the distance.

12. The method as claimed in claim 8, wherein the cost function is equal to the sum of the different terms relative to a measurement point.

13. The method as claimed in claim 12, wherein the cost function fc, for a common reflectivity zone Rc in question, is written thus: $\begin{array}{cc}\mathrm{fc}=\underset{k=1}{\overset{N}{.Math.}}.Math.{\mathrm{Id}}_k-\frac{\mathrm{Rc}\cos {\theta }_k}{d_k^2}.Math.& \left[\mathrm{Math}.3\right]\end{array}$ where Id.sub.k is the diffuse intensity and θ.sub.k is the angle of incidence for a measurement point in question with the index k belonging to the common reflectivity zone in question, and N is the number of measurement points belonging to the common reflectivity zone in question.

14. A computer program including instructions for implementing the method according to claim 8, when the program is executed by a processor.

15. A non-transitory computer-readable recording medium on which is recorded a program for implementing the method according to claim 14, when this program is executed by a computer.

16. The computer as claimed in claim 4, wherein the cost function fc, for a common reflectivity zone Rc in question, is written thus: $\begin{array}{cc}\mathrm{fc}=\underset{k=1}{\overset{N}{.Math.}}.Math.{\mathrm{Id}}_k-\frac{\mathrm{Rc}\cos {\theta }_k}{d_k^2}.Math.& \left[\mathrm{Math}.3\right]\end{array}$ where Id.sub.k is the diffuse intensity, d.sub.k is the distance and θ.sub.k is the angle of incidence for a measurement point in question with the index k belonging to the common reflectivity zone in question, and N is the number of measurement points belonging to the common reflectivity zone in question.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Other features, details and advantages will become apparent from reading the following detailed description and from analyzing the appended drawings, in which:

[0036] FIG. 1 shows different steps of a method for processing the data provided by a lidar according to one embodiment.

[0037] FIG. 2 is a schematic illustration of a common reflectivity zone.

[0038] FIG. 3 shows an example of implementation of the method for processing the data provided by a lidar according to one embodiment.

DESCRIPTION OF THE EMBODIMENTS

[0039] The drawings and descriptions below essentially contain elements of definite character. Consequently they can be used not only to clarify the understanding of the present disclosure, but also to contribute to its definition if necessary.

[0040] FIG. 1 shows different steps of a method for processing the data provided by a lidar, implemented by a computer, according to one embodiment.

[0041] The method comprises a step S10 of obtaining, for a plurality of measurement points of a lidar, data representative of a diffuse intensity Id at a measurement point and a distance d between the measurement point and the lidar.

[0042] As described above in the introductory section, the lidar measures the intensity Ir reflected by a measurement point located on a reflective surface, and the travel time of a light pulse reflected at the measurement point. On the basis of these measurement data, the computer of the lidar can determine the distance d between each measurement point and the lidar, together with the ambient intensity Ia, and the diffuse intensity Id, by subtracting the ambient intensity Ia from the reflected intensity Ir. Since the ambient light intensity Ia is constant for all the measurement points, it can be determined as the continuous component of the reflected intensity Ir measured by the lidar. The ambient intensity Ia is found in a known manner by acquiring the road scene, using the lidar, without illuminating it. The computer is also capable of determining the three-dimensional coordinates of each measurement point, on the basis of the distance d between each measurement point and the lidar and the value of the angle of the beam reflected by the measurement point and received by the lidar.

[0043] Depending on the method of processing the data produced by the lidar, the data representative of a diffuse intensity Id obtained by the computer may be the diffuse intensity Id, the reflected intensity Ir or, possibly, the ambient intensity Ia. Similarly, the data representative of the distance may be directly equal to the distance d determined by the lidar, the measured travel time, or the three-dimensional coordinates (x,y,z) of the measurement points determined by the lidar.

[0044] In one embodiment, the data representative of a diffuse intensity Id at a measurement point and a distance d between the measurement point and the lidar are received by the computer, if the computer implementing the method is separate from the computer of the lidar.

[0045] In another embodiment, the whole of the method described here is implemented by the computer of the lidar. The step of obtaining, S10, is then equivalent to the determination of the diffuse intensity Id and the distance d between the measurement point in question and the lidar.

[0046] In a particular embodiment, the data representative of a diffuse intensity Id and of a distance d between the measurement point and the lidar are obtained by receiving an image and a data structure.

[0047] The image comprises the measurement of the reflected intensity Ir for different measurement points, and the data structure comprises the three-dimensional coordinates (x,y,z) of each measurement point. The distance d between the measurement point having the three-dimensional coordinates (x,y,z) and the lidar is then determined by means of the following formula:

d=√{square root over ((x.sup.2+y.sup.2+z.sup.2))}  [Math. 4]

[0048] Then, in a step S20, the computer determines the direction normal to the surface tangential to each measurement point, and deduces therefrom an angle of incidence θ of the lidar at each measurement point, given the direction of the light beam emitted by the lidar and guided toward the measurement point in question, as described above in the introductory section, for example.

[0049] In a step S30, the computer calculates, for each measurement point, the reflectivity R.sub.d from the angle of incidence θ determined in step S20, the distance d, and the diffuse intensity Id. As described above, the distance d and the diffuse intensity Id are either obtained directly by the computer in step S10 or determined from data obtained in a supplementary step if the computer implementing the method is different from that of the lidar.

[0050] In one embodiment, the reflectivity Rd is calculated by means of the following function for calculating the diffuse intensity:

[00004]$\begin{array}{cc}\mathrm{Id}=\frac{Rd\cos \theta }{d^2}& \left[\mathrm{Math}.2\right]\end{array}$

[0051] In another embodiment, the reflectivity Rd is calculated by means of the following function for calculating the diffuse intensity:

[00005]$\begin{array}{cc}\mathrm{Id}=\frac{\mathrm{Rd}\sin \left(\theta +\frac{\pi }{2}\right)}{d^2}& \left[\mathrm{Math}.5\right]\end{array}$

[0052] where Rd is the reflectivity of the material forming the surface on which the light pulse is reflected, θ is the angle of incidence of the light beam emitted by the lidar onto the point on the surface in question, determined in step S20, and d is the distance between the lidar and the measurement point in question.

[0053] In a step S40, spatially adjacent measurement points having a reflectivity within a given range of values centered on a common reflectivity value Rc are identified in order to form common reflectivity zones.

[0054] An example of a common reflectivity zone is shown in FIG. 2. In this figure, measurement points P1, P2, P3, P4, P5, P6, P7, P8 have coordinates (x1, y1, z1), (x2, y2, z2), (x3, y3, z3), (x4, y4, z4), (x5, y5, z5), (x6, y6, z6), (x7, y7, z7), (x8, y8, z8), are located at a distance d1, d2, d3, d4, d5, d6, d7, d8 from the lidar and have a reflectivity of Rd1, Rd2, Rd3, Rd4, Rd5, Rd6, Rd7, Rd8, respectively.

[0055] As shown in FIG. 2, the measurement points P1, P2, P3, P4, P5 are spatially adjacent, taking into account their coordinates (x1, y1, z1), (x2, y2, z2), (x3, y3, z3), (x4, y4, z4), (x5, y5, z5). It will be noted that the distances d1, d2, d3, d4, d5 associated with these points are not taken into account, since points located at the same distance from the lidar are not necessarily spatially adjacent. It will be noted that the points P1, P2, P3, P4, P5 have a reflectivity of Rd1, Rd2, R3, Rd4 and Rd5, lying within the range of values Rc1±Δ, where, for example, Rc1 is a common reflectivity value and Δis a predetermined value. In particular, the points P1, P2, P3, P4, P5 therefore belong to a zone Z1 having a common reflectivity Rc1.

[0056] In a step S50, a cost function is defined for at least one common reflectivity zone, for example the zone Z1. The cost function comprises a plurality of terms Tk. Each term is relative to a measurement point, in this case P1, P2, P3, P4, P5, and comprises a first diffuse intensity Id.sub.k, in this case Id1, Id2, Id3, Id4, Id5, corresponding to the diffuse intensity determined from the reflected intensity Ir measured by the lidar and the ambient intensity Ia, and a second, theoretical, diffuse intensity determined by means of the diffuse intensity calculation function as defined above [Math. 2] or [Math. 5].

[0057] In particular, the value of the theoretical intensity is determined as a function of the common reflectivity value Rc, in this case Rc1, of the angle θ.sub.k and the distance d.sub.k of the measurement point in question identified here by the index k, where k belongs to {1,. . .,5} in the example of FIG. 2.

[0058] In one embodiment,

[00006]$\begin{array}{cc}\mathrm{Tk}=.Math.{\mathrm{Id}}_k-\frac{\mathrm{Rc}\cos {\theta }_k}{d_k^2}.Math.& \left[\mathrm{Math}.6\right]\end{array}$

[0059] when the diffuse intensity calculation function [Math. 2] is used.

[0060] Thus, the value of reflectivity taken into account for each of the measurement points is not equal to the reflectivity Rd.sub.k calculated in step S30, but is equal to the reflectivity Rc for the common reflectivity zone in question.

[0061] In a particular embodiment, the cost function is equal to the sum of the different terms relative to a measurement point Tk, in this case P1, P2, P3, P4, P5.

[0062] Thus the cost function fc is written:

[00007]$\begin{array}{cc}\mathrm{fc}=\underset{k=1}{\overset{N}{.Math.}}\mathrm{Tk}& \left[\mathrm{Math}.7\right]\end{array}$

[0063] The index k is used to identify the measurement points belonging to the common reflectivity zone in question, and N is equal to the number of measurement points belonging to the common reflectivity zone in question.

[0064] In one embodiment, the cost function fc, when the function [Math. 2] is used, takes the form of:

[00008]$\begin{array}{cc}\mathrm{fc}=\underset{k=1}{\overset{N}{.Math.}}.Math.{\mathrm{Id}}_k-\frac{\mathrm{Rc}\cos {\theta }_k}{d_k^2}.Math.& \left[\mathrm{Math}.3\right]\end{array}$

[0065] In the example considered here, for the zone Z1, Rc=Rc1 and N=5.

[0066] In step S60, the cost function fc is minimized, using an iterative non-linear optimization algorithm, in order to update at least one of the angle θ.sub.k, the distance d.sub.k and the reflectivity Rd.sub.k at the measurement point, for each of the measurement points Pk belonging to a common reflectivity zone, in this case the measurement points P1, P2, P3, P4, P5 having a common reflectivity Rc1.

[0067] It will be noted that, when the reflectivity Rd.sub.k is optimized, the initial value, at the start of the optimization, is the common reflectivity Rc, but that this value is updated during the optimization. If the reflectivity is not optimized, the value Rc remains unchanged.

[0068] Consequently, on completion of step S60, we obtain updated, and therefore more precise, values of the angle of incidence θ.sub.k, the distance d.sub.k and/or the reflectivity Rd.sub.k for each of the measurement points Pk considered.

[0069] In one embodiment, the measurement points belonging to common reflectivity zones also have angles of incidence within a range of permitted values.

[0070] This range of permitted values may be used to constrain the values of the angle of incidence used by the iterative non-linear optimization algorithm. The non-linear optimization algorithm thus converges more rapidly.

[0071] FIG. 3 shows schematically a motor vehicle 1 comprising a lidar 2. As discussed above, the lidar 2 is configured for emitting a light beam comprising a light pulse and for measuring the time of flight and therefore the distance between a measurement point and the lidar, together with the intensity reflected by the measurement point, using a suitable sensor. The lidar 2 therefore comprises a computer 3 configured for processing the data supplied by the sensor and for controlling the light beam so as to determine the distance d and the reflected intensity Ir, or even the ambient intensity Ia and the diffuse intensity Id, for each of the measurement points. The computer 3 is also configured for determining the angle of the beam reflected by the measurement point and received by the lidar for each measurement point, and for determining the three-dimensional coordinates of the measurement points.

[0072] In one embodiment, the method described with reference to FIG. 1 is implemented by the computer 3 of the lidar 2. Thus, in step S10, the data representative of the diffuse intensity Id at the measurement point and of the distance d between the measurement point and the lidar are determined by the computer 3, and stored in memory for use in the implementation of steps S20 to S60. The three-dimensional coordinates (x,y,z) of each measurement point are also determined by the computer 3, and are stored in memory for re-use in the implementation of the method.

[0073] In one embodiment, shown in broken lines in FIG. 3, the method described with reference to FIG. 1 is implemented by a computer 10 different from the computer 3 present in the lidar. In this case, in step S10, the data representative of the diffuse intensity Id at the measurement point and of the distance d between the measurement point and the lidar, determined by the computer 3 for a plurality of measurement points of the lidar, are transmitted to the computer 10. In a preferred embodiment, the intensity Ir measured for each measurement point by the sensor, together with the three-dimensional coordinates of the corresponding measurement points, are then transmitted to the computer 10.

INDUSTRIAL APPLICATION

[0074] The data supplied by lidars are used in many applications, notably in the fields of motor vehicles and aerial imaging. In motor vehicle field, these data may be used for the identification of different types of object on the basis of the reflectivity, and, if appropriate, of the angle of incidence, as well as their distance, by driving assistance systems commonly referred to as ADAS (“Advanced Driver Assistance Systems”). This is particularly useful for applications such as emergency braking, vehicle monitoring or line change detection, for example.

[0075] The present disclosure is not limited to the various embodiments described above purely by way of example, but incorporates all variants that may be envisaged by those skilled in the art in the desired scope of protection.