METHOD AND RADAR SENSOR FOR MONITORING THE TRAFFIC FLOW

Abstract

A method for monitoring the traffic flow with the aid of a stationary radar sensor, which operates in the frequency band of 76 GHz to 77 GHz and is operated in such a way that it meets a radiation protection criterion according to which, in a sequence of consecutive time intervals which all have the same length T, each of these time intervals includes an uninterrupted radiation-free time period of at least 0.9 T. An arbitrary stationary point in the surroundings of the radar sensor within each of these time intervals is exposed to the radar radiation of this radar sensor for no more than a radiation duration of L≤T/100. The radar sensor is static during the measuring operation, and the radiation protection criterion is met by the temporally clocked operation of the radar sensor.

Claims

1. A method for monitoring the traffic using a stationary radar sensor, the method comprising: operating the radar sensor in a frequency band of 76 GHz to 77 GHz in such a way that it meets a radiation protection criterion according to which, in a sequence of consecutive time intervals which all have the same length T, each of the time intervals includes an uninterrupted radiation-free time period of at least 0.9 T, and an arbitrary stationary point in surroundings of the radar sensor within each of these time intervals is exposed to the radar radiation of the radar sensor for no more than a radiation duration of L T/100; wherein the radar sensor is static during a measuring operation, and the radiation protection criterion is met by a temporally clocked operation of the radar sensor.

2. The method as recited in claim 1, wherein the radiation duration L is ≤T/200.

3. The method as recited in claim 1, wherein the radiation-free time period is at least 0.95 T.

4. The method as recited in claim 1, wherein the radiation-free period is at least 0.995 T.

5. The method as recited in claim 1, wherein the length T of the time intervals is 250 ms.

6. The method as recited in claim 1, wherein the radar sensor is an FMCW radar.

7. The method as recited in claim 6, wherein the radar sensor is a rapid chirp FMCW radar.

8. The method as recited in claim 1, wherein, during each of the time intervals, a value v_i is measured for a velocity v of each located radar target, and a final value for the velocity v is calculated based on the values v_i obtained in the multiple measuring cycles.

9. A radar sensor configure to operate in a frequency band of 76 GHz to 77 GHz and being operated in such a way that each measuring cycle of length T includes an activity phase having a length L, during which radar radiation is sent and received, and an idle phase having a length T−L, during which no radar radiation is sent, wherein the length L of the activity phase is no more than T/100.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0020] FIG. 1 shows a diagram of a radar sensor for monitoring the traffic flow in a traffic infrastructure, including a time diagram for illustrating a clocked operation of the radar sensor.

[0021] FIGS. 2A through 2B show a flowchart for illustrating one example embodiment of a method according to the present invention.

DETAILED DESCRIPTION OF EXAMPLE EMBODIMENTS

[0022] FIG. 1 shows a section of a road 10. A motor vehicle 14 is driving on a roadway 12 of this road. A radar sensor 16 is mounted on a pole at the roadside in a stationary and static manner, whose radar lobe 18 is directed at a reference point of roadway 12 situated at a distance d_0 ahead of the position of radar sensor 16. During a time period in which motor vehicle 14 approaches this reference point in the direction of the arrow, passes this point, and then further approaches radar sensor 16, motor vehicle is located by radar sensor 16, and the distance of the motor vehicle is measured periodically, using a period duration T of 250 ms, for example. In this way, it is possible to determine the point in time at which the motor vehicle passes the reference point. Since vehicles following one another pass this point at different times, it is possible in this way to reliably determine the number of vehicles which negotiate roadway 12 within a time period of, for example, one hour or one day, even in the case of high traffic density.

[0023] In the shown example, radar sensor 16 is a rapid chirp FMCW radar which, during each time interval of length T within an activity phase of length L, sends a radar signal whose frequency is modulated in the form of a sequence of short and steep frequency ramps 20. In FIG. 1, frequency f of the sent signal is plotted in a window 22 as a function of time t. As is customary with a rapid chirp FMCW radar, the time signal recorded by the radar sensor during the activity phase is subjected to a two-dimensional Fourier transform (2D FFT). One dimension is used to determine the distance of the radar target, i.e., e.g., of motor vehicle 14, based on the signal profile on each individual frequency ramp 20. The other dimension is used to determine the relative velocity of the located vehicle based on the change in the signal from ramp to ramp.

[0024] In contrast to customary radar sensors of this type, length L of the activity phases here, however, is very small in relation to length T of the entire measuring cycle, so that the majority of the measuring cycle is made up of an idle phase of length T−L, during which no radar signal is sent. For example, the activity phase has length L=1.25 ms, i.e., only 1/200 of length T of the entire measuring cycle. In this way, it is ensured that, during each time interval of length T (250 ms), radiation duration L, during which an on-board radar sensor of motor vehicle 14 is exposed to this radiation, is no more than 1.25 ms. In this way, the extent of interferences between the radiation of radar sensor 16 and the radiation of the on-board radar sensor of motor vehicle 14 is limited to a tolerable degree. During the entire idle phase having length T−L, the radar sensor of motor vehicle 14 may operate completely without impairment so that the traffic situation may be reliably tracked in real time, and a driver assistance system or collision warning system of motor vehicle 14 receives reliable pieces of information and is able to make the necessary decisions quickly and on a reliable basis.

[0025] FIG. 2 shows one example of a method, in the form of a flowchart, with the aid of which, using radar sensor 16, a traffic count may be carried out, a velocity statistic of the vehicles driving on roadway 12 may be created and, in the case of a traffic jam, a traffic jam alert may be generated.

[0026] In step S1, a radar signal is sent during an activity phase of length L, whose frequency f is in the frequency band of 76 GHz to 77 GHz. The frequency is modulated in the form of a sequence of frequency ramps 20. The radar echoes received by the radar targets are received, digitized, and recorded as a time signal over the duration of the activity phase. The signals recorded on the various frequency ramps form a two-dimensional data field in which one dimension indicates the time within the respective frequency ramp, and the second dimension identifies the frequency ramp.

[0027] In step S2, a two-dimensional spectrum is then formed by a two-dimensional Fourier transform (2D FFT), in which each located object, for example motor vehicle 14, becomes apparent as a peak. The position of this peak in the first dimension indicates measured distance d i of this object, and the position of the peak in the second dimension indicates (with lesser accuracy) velocity v_i of this object. Values d_i and v_i are stored.

[0028] In step S3, it is checked whether measured distance d i is smaller than distance d_0. If this is not the case, i.e., when vehicle 14 has not yet passed the reference point, a jump is made to step S4 and, after idle time T−L has elapsed, a jump is made back to step S1, and a new measuring cycle begins.

[0029] If, in step S3, the result is “yes,” it is checked in step S5 whether a counting index i has the value 0. If this is the case, the located object is classified in step S6 to determine the vehicle type in greater detail. For example, a distinction may be made between trucks, passenger cars, and two-wheelers using an angle-resolving radar sensor based on the measured object dimensions. The different vehicle types are counted separately in step S6. If the object was classified as a passenger car, the counter content for passenger cars is increased by 1, and a corresponding procedure is carried out for the other vehicle types. In addition, counting index i is initialized to the value 1 in step S6. Thereafter, a jump takes place back to step S4.

[0030] If it was established in step S5 that counting index i is already different from 0, this index is increased by 1 in step S7, and it is checked in step S8 whether the index has already reached a certain value N. If this is not the case, a jump takes place back to step S4. Otherwise, counting index i is reset to 0 in step S9, and a jump takes place to step S10 in FIG. 2B.

[0031] The routine described up to here ensures that a located object, such as, for example, vehicle 14, is tracked over N measuring cycles starting at the point in time at which it has passed the reference point, an updated value d_i for the distance and an updated value v_i for the velocity being stored during each measuring cycle.

[0032] In step S10, a more precise value for velocity v of the located object is calculated based on these stored values. The calculation may take place, for example, by averaging over stored values v_i or by calculating difference d_0−_(N−1) and division by N*T (the overall duration of N measuring cycles). Optionally, it is also possible to carry out both calculation methods in parallel, and to then form a weighted sum from the results.

[0033] In step S11, a velocity statistic is updated based on velocity v calculated in step S10, which indicates the velocity distribution of the located vehicles.

[0034] In step S12, it is checked whether the measured velocity v is smaller than a certain threshold value, for example 5 m/s. If this is the case, this means that the traffic is backing up on roadway 12, and the most recently located vehicle is stationary or only driving at walking speed. In this case, a traffic jam alert is sent in step S13. Thereafter, a jump takes place back to step S4 in FIG. 2A. If the result in step S12 is negative, step S13 is skipped.

[0035] In the next measuring cycle, counting index i has the value 0, and it is waited until the next vehicle has reached the reference point at d_0. If the number N of the measuring cycles per vehicle is 3 or 4, for example, the overall measuring duration per vehicle is smaller than the smallest temporal distance to be expected between consecutive vehicles. Optionally, however, the method may also be expanded and modified in such a way that multiple vehicles located shortly after one another may be tracked in parallel. In this case, a longer overall measuring duration per vehicle is possible, so that more precise velocity values are obtained.