Estimation device and display device

Abstract

To provide an estimation device for increasing estimation precision of an existence probability.SOLUTION: An estimation device 100 comprises an existence probability estimation section P2 for estimating an existence probability of an object in an estimated area ETA around a vehicle 1 on the basis of sensor data including a detection point DP in which the periphery of the vehicle 1 is detected. The existence probability estimation section P2 sets mutually different existence probabilities among a detection area A1 corresponding to the detection point DP of current sensor data, a proximity area A2 close to the detection area A1, and a remaining area A3 excluding the detection area A1 and the proximity area A2 in the estimated area ETA.SELECTED DRAWING: Figure 8

Description

[Technical Field]

[0001] The disclosure of this specification relates to a technology for estimating the probability of an object existing in an estimation target area around a moving object. [Background technology]

[0002] Patent Document 1 discloses compressing an OGM (Occupancy Grid Map) that maps the probability of an object existing in an estimation target area around a moving object. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2020-42726 Summary of the Invention [Problem to be solved by the invention]

[0004] Compressing the estimation results as in Patent Document 1 makes them easier to use in moving objects. However, if the estimation accuracy of the existence probability is poor to begin with, there is a possibility that the presence of an object will be falsely detected.

[0005] One of the purposes of the disclosure of this specification is to provide an estimation device with improved accuracy in estimating the existence probability, and a display device capable of displaying the estimation results obtained by this estimation device. [Means for solving the problem]

[0006] One aspect disclosed herein is an estimation device including a presence probability estimation unit (P2) that estimates the presence probability of an object in an estimation target area (ETA) around a moving object (1) based on sensor data including detection points (DP) that detect the periphery of the moving object (1), The existence probability estimation unit sets different existence probabilities among a detection area (A1) corresponding to the detection point of the current sensor data, a nearby area (A2) close to the detection area, and a remaining area (A3) excluding the detection area and the nearby area, within the estimation target area.

[0007] According to this aspect, instead of simply setting different existence probabilities for detection points and non-detection points, spatial continuity is taken into consideration and individual existence probabilities are set for areas close to the detection points where the possibility of an object being present is considered to be higher than in areas far from the detection points. Setting the existence probabilities with a focus on spatial continuity makes it possible to improve the estimation accuracy of the existence probabilities.

[0008] Another disclosed aspect is a display device communicably connected to the estimation device, a display content generating unit (32a) that generates display content by imaging the existence probability map; and a display (32b) having a screen for displaying the display content.

[0009] According to this aspect, the displayed existence probability map sets the existence probability by focusing on spatial continuity, and therefore it is possible to provide the viewer with an accurate estimate of the map while reducing the sense of incongruity of the map.

[0010] Note that the symbols in parentheses included in the claims etc. are intended to exemplify the correspondence with the parts of the embodiments described below, and are not intended to limit the technical scope. [Brief explanation of the drawings]

[0011] [Figure 1] A top view showing a millimeter-wave radar mounted on a vehicle and its detection area. [Figure 2] A top view showing a millimeter-wave radar mounted on a vehicle and its detection area. [Figure 3] FIG. 1 is a diagram illustrating a schematic configuration of an estimation device and a display device. [Figure 4]10 is a flowchart showing an example of processing by the estimation device. [Figure 5] 10 is a flowchart showing an example of processing by the estimation device. [Figure 6] FIG. 10 is a diagram showing a previous existence probability map. [Figure 7] A diagram showing the detection results. [Figure 8] A diagram showing the current probability addition value map. [Figure 9] A diagram showing the current existence probability map. [Figure 10] FIG. 1 is a diagram illustrating a schematic configuration of an estimation device and a display device. [Figure 11] 10 is a flowchart showing an example of processing by the estimation device. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, several embodiments will be described with reference to the drawings. Note that corresponding components in each embodiment are given the same reference numerals, and redundant description may be omitted. When only a portion of the configuration is described in each embodiment, the configuration of another embodiment described previously can be applied to the remaining portion of the configuration. Furthermore, in addition to the combinations of configurations explicitly stated in the description of each embodiment, configurations of several embodiments can also be partially combined together even if not explicitly stated, as long as there is no particular problem with the combination.

[0013] (First embodiment) The estimation device 100 of this embodiment is mounted on a vehicle 1, which is a moving body, as shown in FIGS. 1 and 2. The vehicle 1 is equipped with a millimeter-wave radar 11 as a periphery monitoring sensor for detecting objects in a peripheral monitoring area DA. As shown in FIG. 1, a total of one millimeter-wave radar 11 may be mounted on the vehicle 1. In this example, the millimeter-wave radar 11 is mounted on the front bumper of the vehicle 1 and detects a monitoring area DA in front of the vehicle 1. As shown in FIG. 2, a plurality of millimeter-wave radars 11 may be mounted on the vehicle 1. In this example, the plurality of millimeter-wave radars 11 are mounted on the front and rear bumpers of the vehicle 1 and detect monitoring areas DA in front of, on the front sides of, and on the rear sides of the vehicle 1 that are offset from one another.

[0014] The millimeter-wave radar 11 may employ, for example, a frequency modulation continuous wave (FMWC) system. The millimeter-wave radar 11 emits millimeter waves or quasi-millimeter waves (transmitted waves) from a transmitting antenna toward a monitoring area DA. The millimeter-wave radar 11 then receives the waves reflected by moving or stationary objects, etc., at a receiving antenna. The transmitted waves and received waves are mixed in a mixer to generate an IF (intermediate frequency) signal. The IF signal is processed by fast Fourier transform or the like to obtain the position where the transmitted waves are reflected by an object (hereinafter referred to as the detection point) and speed information.

[0015] 3, the millimeter-wave radar 11 can provide the thus obtained detection point and speed information to the estimation device 100 as sensor data. The sensor data may include information on the intensity of the reflected wave, or may include information on the spectral spread of the reflected peak obtained by converting the reflected wave into frequency-domain information by fast Fourier transform. The spectral spread may be represented by a half-width based on the intensity of the reflected peak, or may be represented by another index.

[0016] The estimation device 100 is a device that estimates the presence probability of an object in an estimation target area around the vehicle 1 based on sensor data including detection points. The presence probability here may be a numerical value expressed in the range of 0 to 1, or may be a numerical value expressed as a percentage. As will be described in detail later, the presence probability may be expressed in the form of logarithmic odds by taking a logarithm. In other words, the presence probability may be expressed in any form as long as it is a numerical value indicating the presence probability. The estimation device 100 is connected to a millimeter-wave radar 11, a vehicle speed sensor 21, a yaw rate sensor 22, a driving assistance system 31, a display device 32, etc., via a wire harness, an in-vehicle network such as a CAN (Control Area Network (registered trademark)), etc., so as to be able to communicate with them via wire or wirelessly.

[0017] The estimation device 100 may be, for example, an electronic control unit (ECU) mounted on the vehicle 1, or may be configured mainly by a computer. The computer configuring the estimation device 100 has at least one memory and one processor. The memory may be at least one type of non-transient tangible storage medium, such as a semiconductor memory, a magnetic medium, or an optical medium, that non-temporarily stores computer programs and data that can be read by the processor. Furthermore, a rewritable volatile storage medium, such as a random access memory (RAM), may be provided as the memory. The processor may include, as a core, at least one type of core, such as a central processing unit (CPU), a graphics processing unit (GPU), or a reduced instruction set computer (RISC)-CPU.

[0018] The dedicated computer constituting the main unit 51 may be an SoC (System on a Chip) that integrates memory, a processor, and an interface into a single chip, or may have at least one SoC as a component of the dedicated computer.

[0019] The dedicated computer may further include circuits such as a field-programmable gate array (FPGA) or an application-specific integrated circuit (ASIC), which may be configured to perform a specific portion of the processing of the estimation device 100.

[0020] The estimation device 100 includes an egomotion calculation unit P1, an existence probability estimation unit P2, and an object existence determination unit P3 as processing units for implementing processing by a processor executing a computer program or the like.

[0021] The egomotion calculation unit P1 calculates the amount of movement of the vehicle 1. The amount of movement of the vehicle 1 may be the amount of movement of the vehicle 1 per unit time, and may include the distance and direction of movement. The calculation of the amount of movement of the vehicle 1 may be performed based on the vehicle speed sensor 21 and yaw rate sensor 22 of the vehicle 1. The vehicle speed sensor 21 detects the speed of the vehicle 1. The yaw rate sensor 22 detects the yaw rate (angular velocity) acting on the vehicle 1.

[0022] The presence probability estimation unit P2 repeatedly executes a calculation to estimate the presence probability of an object based on the sensor data. Specifically, the presence probability estimation unit P2 generates an OGM (hereinafter referred to as a presence probability map) that is a region included in the monitoring area DA and that is an estimation target region around the vehicle 1, and that is a two-dimensional map of the presence probability of an object. The estimation target region is set in advance as a region of any shape, such as a rectangle or a sector. If the estimation target region is rectangular, for example, it has a plurality of grids that are partitioned into a grid based on a Cartesian coordinate system. If the estimation target region is sector-shaped, for example, it has a plurality of grids that are partitioned into a grid based on a polar coordinate system.

[0023] The presence probability estimation unit P2 acquires the amount of movement of the vehicle 1 from the egomotion calculation unit P1 and reflects the amount of movement in the estimation calculation. As a result, the estimation target area becomes an area that moves together with the vehicle 1 and maintains a specific relative positional relationship with the vehicle 1.

[0024] The object presence determination unit P3 determines the presence of an object in the estimation target area based on the presence probability map generated by the presence probability estimation unit P2. For example, the object presence determination unit P3 may be configured to determine that an object exists in a grid where the presence probability is equal to or greater than a preset presence determination threshold, and to determine that an object does not exist in a grid where the presence probability is less than the presence determination threshold. Note that, for example, when the presence probability is expressed in a range of 0 to 1, the presence determination threshold may be 0.4, 0.5, or 0.6. A grid where it is determined that an object exists is designated as an object presence area. The currently generated presence probability map and information on the object presence area may be stored in memory and used for the next estimation calculation.

[0025] Next, an example of a processing method by the estimation device 100 will be described using the flowchart in Fig. 4. The series of processes from S10 to S40 may be realized, for example, by the processor of the estimation device 100 executing a computer program stored in memory. This series of processes is repeatedly executed, for example, periodically.

[0026] First, in S10, the presence probability estimation unit P2 acquires the latest sensor data from the millimeter wave radar 11. At the same time, the presence probability estimation unit P2 acquires the previous presence probability map. After processing S10, the process proceeds to S20. In S20, the egomotion calculation unit P1 calculates the amount of movement of the vehicle 1. After processing S20, the process proceeds to S30.

[0027] In S30, the presence probability estimation unit P2 executes a presence probability determination process based on the sensor data, the previous presence probability map, and the amount of movement of the vehicle 1. As a result, a current presence probability map is generated. After the process of S30, the process proceeds to S40.

[0028] In S40, the object presence determination unit P3 determines the presence of an object in the estimation target area based on the current presence probability map. As a result, information on the object presence area is generated. The series of processes ends with S40.

[0029] The existence probability determination process (processing for generating an existence probability map) of S30 will be described in detail using the flowchart of Fig. 5. In the first step S301, the existence probability estimation unit P2 determines whether or not calculation has been completed for the entire region (all grids) of the estimation target region. If Yes, the series of processes ends. If No, proceed to S302.

[0030] S302 to S306 are processes for one grid that has not yet been calculated. That is, at the start of S302, the existence probability estimation unit P2 selects one grid that has not yet been calculated as the calculation target. Then, the existence probability estimation unit P2 acquires the current (latest) detection point from the sensor data. The existence probability estimation unit P2 compares the acquired detection point with the grid that is the calculation target. In S302, the existence probability estimation unit P2 determines whether or not there is a detection point whose position corresponds to the grid that is the calculation target. If the answer is Yes, it is determined that the grid that is the calculation target belongs to the detection area that corresponds to the detection point of the current sensor data, and the process proceeds to S304. If the answer is No, the process proceeds to S303.

[0031] In S303, the existence probability estimation unit P2 determines whether or not a detection point exists within a predetermined distance of the grid to be calculated. If the result is Yes, it is determined that the grid to be calculated belongs to a nearby region that is close to the detection region, and the process proceeds to S305. If the result is No, it is determined that the grid to be calculated belongs to the remaining region of the estimation target region excluding the detection region and the nearby region, and the process proceeds to S306.

[0032] In S304, the existence probability estimation unit P2 executes detection region probability processing on the grids to be calculated that are determined to belong to the detection region. After the processing of S304, the process returns to S301.

[0033] In S305, the existence probability estimation unit P2 executes proximity region probability processing on the grids to be calculated that are determined to belong to the proximity region. After the processing of S305, the process returns to S301.

[0034] In S306, the existence probability estimation unit P2 executes a remaining region probability process for the grids to be calculated that are determined to belong to the remaining region. After the process of S306, the process returns to S301.

[0035] In this way, the presence probability estimation unit P2 classifies the estimation target area into a detection area, a proximity area, and a remaining area, and performs processing related to the presence probabilities that differ from one another. In these processing steps, temporal continuity is taken into account by performing probability change operations such as increasing or decreasing the previously generated presence probability map based on the position of the current detection point. Furthermore, by introducing processing for the proximity area, these processing steps take into account spatial continuity based on the positional relationship with the detection point. As a result, the presence probabilities are set to be different for each area.

[0036] The millimeter-wave radar 11 uses a relatively long wavelength, making it difficult to obtain diffuse reflection. Therefore, the position of the detection point is likely to fluctuate depending on the positional relationship between the millimeter-wave radar 11 and the object. For this reason, the millimeter-wave radar 11 in particular is highly effective in improving estimation accuracy by taking into account temporal continuity and spatial continuity.

[0037] Here, we will explain the formula for the probability change operation that can be used commonly for each area. The probability addition value logarithmic odds l(i,t) at time t, which corresponds to the amount of change this time from the previous existence probability, is expressed by the formula set forth in the following equation 1. Here, P(i,t) is the probability addition value this time (time t). i is the index of the grid, and the current grid and the previous grid, which are associated with each other by referring to the amount of movement of vehicle 1 between time t-1 and time t, can be designated with the same index.

[0038]

number

[0039]

number

[0040]

number

[0041] Specifically, for the detection region, the presence probability estimation unit P2 increases the current presence probability of the calculation target grid from the presence probability of the previous grid corresponding to the grid. Note that if the previous presence probability of the detection region is 1 or a preset upper limit, it is not desirable to increase the presence probability, and the previous presence probability may be maintained. For the remaining region, the presence probability estimation unit P2 decreases the current presence probability of the calculation target grid from the presence probability of the previous grid corresponding to the grid. Note that if the previous presence probability of the remaining region is 0 or a preset lower limit, it is not desirable to decrease the presence probability, and the previous presence probability may be maintained. For the adjacent region, the presence probability estimation unit P2 controls the amount of change in the current presence probability from the previous presence probability between the amount of change in the detection region and the amount of change in the remaining region. For example, if the amount of change in the detection region is +0.2 and the amount of change in the remaining region is -0.2, the amount of change in the adjacent region is set to a range from -0.2 to +0.2, or to a change greater than -0.2 and less than +0.2. By applying intermediate processing to the adjacent region, even if a change occurs in the detection point, it is possible to prevent the existence probability at the previous detection point from decreasing too much.

[0042] For example, the presence probability estimation unit P2 may slightly increase or decrease the current presence probability for the nearby region from the previous presence probability. As an example of this, the presence probability estimation unit P2 may set the direction of change for the nearby region depending on whether the presence probability of the previous grid corresponding to the grid to be calculated is equal to or greater than the presence determination threshold.

[0043] When the previous existence probability is equal to or greater than the existence determination threshold, the existence probability estimation unit P2 may perform an operation so that the current existence probability is equal to or greater than the existence determination threshold and has changed downward from the previous existence probability.When the previous existence probability is less than the existence determination threshold, the existence probability estimation unit P2 may perform an operation so that the current existence probability is less than the existence determination threshold and has changed upward from the previous existence probability.In this way, by performing an operation so that the existence probability of the nearby region does not cross the existence determination threshold, it is possible to respond to fluctuations in the detection point within a range that does not overturn the previous existence determination result.

[0044] For example, the presence probability estimation unit P2 may maintain the current presence probability for the nearby region at the previous presence probability, i.e., P(i,t) = 0. Furthermore, the determination of whether to maintain the presence probability for the nearby region may be determined based on whether a probability change operation has been performed within a predetermined period in a grid with the same index as the calculation target grid. For a region in which a probability change operation has been performed within a predetermined period, the presence probability estimation unit P2 may maintain the current presence probability at the previous presence probability. For a region in which a probability change operation has not been performed within a predetermined period, the presence probability estimation unit P2 may change the current presence probability from the previous presence probability to a lower value. This can prevent the presence probability of an undetected region from remaining high for a long period of time.

[0045] Furthermore, the presence probability estimation unit P2 may set a value greater than 0 as the lower limit of the presence probability of the estimation target area. This is because it is inappropriate for the estimation device 100 to determine that an object does not exist, since abnormalities may occur in the detection of the perimeter monitoring sensor. In this case, the lower limit should be set so that a single increase in the probability change operation can cause the current presence probability to exceed the presence determination threshold.

[0046] Next, an example of generating an existence probability map will be described with reference to Figures 6 to 10. Figure 6 shows (1) the previous existence probability map. The existence probability estimation unit P2 generates the current existence probability map based on this previous existence probability map and the position of the detection point DP in the latest sensor data.

[0047] Specifically, Fig. 7 shows (2) the current detection result together with its correspondence with the grid. The millimeter-wave radar 11 detects a plurality of detection points DP corresponding to reflected waves reflected by other vehicles OV that cannot actually be recognized by the millimeter-wave radar 11. The presence probability estimation unit P2 classifies the estimation target area ETA into a detection area A1, a proximity area A2, and a remaining area A3 based on the positional relationship with these detection points DP.

[0048] Fig. 8 shows (3) the probability addition value map. The probability addition values ​​are set to be different for each of the detection area A1, the proximity area A2, and the remaining area A3. In Fig. 8, areas with relatively dark coloring indicate an increasing change, and areas with relatively light coloring indicate a decreasing change.

[0049] Figure 9 shows (4) the current probability existence map, which is the result of applying the probability addition value of (3) the probability addition value map to (1) the previous probability existence map.

[0050] Based on the current probability presence map thus generated, the object presence determination unit P3 determines the presence of an object as described above. Then, the object presence determination unit P3 can output the presence probability map and information on the object presence area to the driving system 31 and display device 32 of the vehicle 1. Information on the range of the proximity area may be added to the presence probability map output here. Specifically, classification information on the detection area, proximity area, and remaining area may be added to each grid. In other words, the object presence determination unit P3 also functions as a map output unit.

[0051] The driving system 31 may sequentially acquire the presence probability map and use the presence probability map to perform automatic driving of the vehicle 1 or to perform driving assistance for the driver of the vehicle 1. The driving assistance may be realized by the driving system 31 executing an ADAS (Advanced Driving Assistant System) application. Examples of the ADAS application include ACC (Adaptive Cruise Control), PCS (Pre-Crash Safety system), and LTA (Lane Tracing Assist).

[0052] The display device 32 is a device that visually presents information to passengers such as a driver in the vehicle 1. The display device 32 includes, for example, a display content generating unit 32a and a display 32b.

[0053] The display content generation unit 32a may be mainly configured with a computer. The computer that constitutes the display content generation unit 32a has at least one memory and one processor. The display content generation unit 32a sequentially acquires the existence probability map and generates display content by visualizing the existence probability map. The display content may be realized by a representation method that distinguishes the existence probability by color shading as shown in FIG. 9, or by other representation methods.

[0054] The display 32b may be a display having a screen for displaying real images, such as a liquid crystal display or an organic EL display. The display 32b may be a meter display arranged opposite the driver's seat in the instrument panel of the vehicle 1, or may be a center information display arranged in the center of the instrument panel. The display 32b displays display content that visualizes the presence probability map sequentially generated by the display content generation unit 32a.

[0055] According to the first embodiment described above, in the estimation device 100, rather than simply setting different existence probabilities for detection points and non-detection points, spatial continuity is taken into consideration and individual existence probabilities are set for areas close to the detection points where the possibility of an object being present is considered to be higher than in areas far from the detection points. Setting the existence probabilities with a focus on spatial continuity makes it possible to improve the estimation accuracy of the existence probabilities.

[0056] Furthermore, according to the first embodiment, the existence probability map displayed on the display device 32 is one in which the existence probability is set with a focus on spatial continuity, and therefore it is possible to provide the viewer viewing the map with an accurate estimation result and reduce the sense of incongruity of the map.

[0057] (Second embodiment) 11 and 12, the second embodiment is a modification of the first embodiment. The second embodiment will be described, focusing on the differences from the first embodiment.

[0058] In the second embodiment, the presence probability estimation unit P2 is set to change the range (size) of the proximity area according to a preset index, which may be at least one of the intensity of the reflected wave, the spread of the spectrum of the reflected wave, the previous presence probability, and object information.

[0059] Specifically, the presence probability estimation unit P2 may determine the range of the proximity area corresponding to a detection point according to the intensity of the reflected wave corresponding to the detection point. The intensity of the reflected wave here may be the value of the power detected by the millimeter-wave radar 11 or the peak intensity in the information on the frequency axis described above. Here, the presence probability estimation unit P2 may expand the range of the proximity area as the intensity increases.

[0060] Furthermore, the presence probability estimation unit P2 may determine the range of the proximity region corresponding to the detection point in accordance with the spread of the spectrum of the reflected wave corresponding to the detection point. Here, the presence probability estimation unit P2 may expand the range of the proximity region as the spread increases. That is, the stronger the reflected wave and the greater the spread of the spectrum, the lower the possibility that the detection point is noise and the higher the possibility that an object exists at or around the detection point.

[0061] Furthermore, the presence probability estimation unit P2 may determine the range of the proximity region corresponding to the current detection point according to the previous presence probability of the region corresponding to the detection point. The presence probability estimation unit P2 may expand the range of the proximity region as the previous presence probability of the region is higher. In other words, the higher the previous presence probability, the lower the possibility that the corresponding detection point is noise and the higher the possibility that an object exists at the detection point and its surroundings.

[0062] Furthermore, the presence probability estimation unit P2 may determine the range of the proximity region corresponding to the detection point in accordance with object information. The object information here may be information about the object presence region determined by the object presence determination unit P3 based on the previous presence probability map. For example, the presence probability estimation unit P2 may determine the range of the proximity region based on whether the region corresponding to the current detection point was designated as the previous object presence region, or may simply determine the range of the proximity region in accordance with the size of the object estimated from the object presence region.

[0063] Furthermore, the object information may be information about the object obtained by a perimeter monitoring sensor other than the millimeter-wave radar 11. For example, as shown in FIG. 10, the presence probability estimation unit P2 may acquire object information from a camera serving as another perimeter monitoring sensor. The object information here may be the image itself of the object captured, or information about the size and type of the object analyzed from the image using a method such as semantic segmentation. The presence probability estimation unit P2 may expand the range of the proximity region when the type of object is estimated to be a dynamic object compared to when the type of object is estimated to be a static object.

[0064] Next, an example of a method for determining the range of the proximity region by the estimation device 100 will be described with reference to the flowchart in Fig. 11. This series of processes from S1001 to S1006 is executed for each detection point, for example, and the results can be reflected in the process of S303 described with reference to Fig. 5.

[0065] In the first step S1001, the existence probability estimation unit P2 determines whether the intensity of the reflected wave is equal to or greater than a preset threshold. If the result is Yes, the process proceeds to S1006. If the result is No, the process proceeds to S1002.

[0066] In S1002, the existence probability estimation unit P2 determines whether the spread of the spectrum of the reflected wave is equal to or greater than a preset threshold. If Yes, proceed to S1006. If No, proceed to S1003.

[0067] In S1003, the presence probability estimation unit P2 determines whether the previous presence probability is equal to or greater than a preset threshold. The threshold here may be the same as or different from the presence determination threshold described above. If the answer is Yes, proceed to S1006. If the answer is No, proceed to S1004.

[0068] In S1004, the existence probability estimation unit P2 determines whether the size of the object is equal to or larger than a preset threshold. If Yes, proceed to S1006. If No, proceed to S1005.

[0069] In S1005, the existence probability estimation unit P2 sets the range of a normal proximity region for the detection point that was the subject of judgment. For example, this may be within the range of the distance of one side of the grid from the detection point. The series of processes ends with S1005.

[0070] In S1006, the existence probability estimation unit P2 sets the range of the proximity region for the detection point that was the subject of the judgment to be expanded more than in S1005. For example, this may be within a range from the detection point to a distance twice the length of one side of the grid. The series of processes ends with S1006.

[0071] (Other embodiments) Although multiple embodiments have been described above, the present disclosure should not be construed as being limited to those embodiments, and can be applied to various embodiments and combinations within the scope that does not deviate from the gist of the present disclosure.

[0072] In another embodiment, the estimation device 100 may be formed integrally with the millimeter-wave radar 11. In yet another embodiment, the estimation device 100 may not include one or both of the egomotion calculation unit P1 and the object presence determination unit P3. The egomotion calculation unit P1 and the object presence determination unit P3 may be mounted on the vehicle 1 as devices separate from the estimation device 100. In yet another embodiment, the estimation device 100 and its functions may be incorporated into the driving system 31 or the display device 32.

[0073] In another embodiment, the estimation device 100 may be configured to estimate the object presence probability based on sensor data from a perimeter monitoring sensor other than the millimeter-wave radar 11. For example, the estimation device 100 may estimate the object presence probability based on sensor data from a LiDAR (Light Detection and Ranging / Laser Imaging Detection and Ranging). The estimation device 100 may also estimate the object presence probability based on sensor data from multiple types of perimeter monitoring sensors.

[0074] In another embodiment, the existence probability map generated by the existence probability estimation unit P2 may be a three-dimensional map instead of a two-dimensional map.

[0075] In another embodiment, the presence probability estimation unit P2 may reflect the amount of movement of a dynamic object in the range of the proximity area based on time-series tracking information of the dynamic object obtained using other surrounding monitoring sensors, V2X communication, etc.

[0076] In another embodiment, the presence probability estimation unit P2 may apply a probability addition value according to the distance from the detection point, rather than applying a uniform probability addition value to the proximity region. For example, the presence probability estimation unit P2 may adopt a probability addition value that decreases the presence probability for a region of the proximity region that is farther away from the detection point.

[0077] The controller and methods described herein may be implemented by a special-purpose computer comprising a processor programmed to perform one or more functions embodied in a computer program. Alternatively, the apparatus and methods described herein may be implemented by special-purpose hardware logic circuitry. Alternatively, the apparatus and methods described herein may be implemented by one or more special-purpose computers comprising a processor executing a computer program in combination with one or more hardware logic circuits. Furthermore, the computer program may be stored as instructions executed by a computer on a computer-readable non-transitory storage medium. [Explanation of symbols]

[0078] 1: vehicle (moving body), 23: display device, 23a: display content generation unit, 23b: display, 100: estimation device, A1: detection area, A2: proximity area, A3: remaining area, DP: detection point, ETA: estimation target area

Claims

1. An estimation device including an existence probability estimation unit (P2) that estimates the existence probability of an object in an estimation target area (ETA) around a moving object (1) based on sensor data including detection points (DP) that detect the periphery of the moving object, The existence probability estimation unit sets different existence probabilities among a detection area (A1) corresponding to the detection point of the current sensor data, a nearby area (A2) close to the detection area, and a remaining area (A3) excluding the detection area and the nearby area, within the estimation target area.

2. The existence probability estimation unit The estimation device according to claim 1 , wherein an amount of change in the current existence probability for the nearby region from the previous existence probability for the nearby region is controlled between an amount of change applied to the detection region and an amount of change applied to the remaining region.

3. The existence probability estimation unit The current presence probability is changed in an increasing direction from the previous presence probability for the detection area; For the remaining region, the current existence probability is changed in a decreasing direction from the previous existence probability; The estimation device according to claim 2 , wherein the current presence probability for the proximity region is controlled between an amount of change in an increasing direction that is smaller than that for the detection region and an amount of change in a decreasing direction that is smaller than that for the remaining region.

4. The estimation device according to claim 2, wherein the existence probability estimation unit performs an operation for the proximity region when the previous existence probability is less than a predetermined existence determination threshold so that the current existence probability is less than the existence determination threshold and has changed in an increasing direction from the previous existence probability.

5. The estimation device according to claim 2, wherein the existence probability estimation unit performs an operation for the nearby region such that, when the previous existence probability is equal to or greater than a predetermined existence determination threshold, the current existence probability is equal to or greater than the existence determination threshold and is a probability that has changed in a decreasing direction from the previous existence probability.

6. The estimation device according to claim 2 , wherein the existence probability estimation unit maintains the previous existence probability as the current existence probability for the adjacent region.

7. The existence probability estimation unit For the area among the adjacent areas in which a probability change operation has been performed within a preset period, the current existence probability is maintained at the previous existence probability; The estimation device according to claim 2 , wherein for an area in which the probability change operation has not been performed within a preset period, the current existence probability is changed in a direction decreasing from the previous existence probability.

8. The estimation device according to claim 1, wherein the existence probability estimation unit sets a lower limit value of the existence probability of the estimation target area so that a single increasing probability change operation can cause the current existence probability to exceed a predetermined existence determination threshold.

9. The sensor data includes the intensity of the reflected wave at the detection point detected by the millimeter wave radar (11), The estimation device according to claim 1 , wherein the presence probability estimation unit determines the range of the proximity region according to the intensity of the reflected wave.

10. The sensor data includes a spectrum of a reflected wave detected at the detection point by a millimeter wave radar (11), The estimation device according to claim 1 , wherein the presence probability estimation unit determines the range of the proximity region in accordance with the spread of the spectrum of the reflected wave.

11. The estimation device according to claim 1 , wherein the presence probability estimation unit determines the range of the proximity region according to a previous presence probability.

12. The estimation device according to claim 1 , wherein the presence probability estimation unit determines the range of the proximity region in accordance with information about the object.

13. The estimation device according to any one of claims 1 to 12, further comprising an object presence determination unit (P3) that determines the presence of the object based on whether the presence probability estimated by the presence probability estimation unit is equal to or greater than a predetermined presence determination threshold.

14. The estimation device according to claim 1 , further comprising a map output unit (P3) that outputs an existence probability map to which information relating to the range of the proximity region is added.

15. A display device communicably connected to the estimation device according to any one of claims 1 to 12, a display content generating unit (32a) that generates display content by imaging the existence probability map; a display (32b) having a screen for displaying the display content.

Images