Recognition device, control device for moving body, recognition method, and program

Abstract

A recognition device for recognizing the position and speed of an object present in the periphery of a moving body on the basis of the output of a detection device for detecting the situation in the periphery of the moving body, the recognition device being provided with: a first recognition unit for repeatedly outputting, at a first cycle, the position of the object, which is the result of performing a process for recognizing the position of the object; and a second recognition unit that repeatedly outputs the speed of the object, which is the result of performing the processing for recognizing the speed of the object, at a second cycle that is shorter than the first cycle.

Description

Technical Field

[0001] This invention relates to an identification device, a control device for a moving body, an identification method, and a procedure. Background Technology

[0002] In recent years, research and practical applications have been advancing towards enabling autonomous movement of vehicles and other mobile bodies. In such technologies, the ability to identify objects surrounding the mobile body (hereinafter referred to as perimeter recognition technology) is essential in almost all cases. As an invention related to perimeter recognition technology, the following invention is disclosed: based on driving scene data and driving behavior data in the current driving scenario of the vehicle, it is determined whether prescribed stable driving conditions are met; if the prescribed stable driving conditions are met, driving scene data and driving behavior data are collected at a frequency lower than a prescribed sampling frequency, thereby reducing data redundancy in similar scenarios and similar driving modes (Patent Document 1).

[0003] Prior art literature

[0004] Patent documents

[0005] Patent Document 1: Japanese Patent Publication No. 2021-526699 Summary of the Invention

[0006] The problem that the invention aims to solve

[0007] In implementing perimeter recognition technology, it is essential to simultaneously reduce processor load and rapidly track emergency situations. While previous technologies comprehensively assessed the environment of moving objects, there are concerns about their inability to quickly track localized changes. Specifically, previous technologies may sometimes fail to rapidly track changes in an object's speed.

[0008] The present invention was made in consideration of such circumstances, and one of its objectives is to provide an identification device, a control device for moving objects, an identification method, and a program that can rapidly follow changes in the speed of objects.

[0009] Solution for solving the problem

[0010] The identification device, the control device for the moving body, the identification method and the program involved in this invention adopt the following structure.

[0011] (1): One aspect of the present invention relates to an identification device that identifies the position and speed of an object existing around a moving body based on the output of a detection device for detecting the surrounding conditions of the moving body, wherein the identification device comprises: a first identification unit that repeatedly outputs the result of processing for identifying the position of the object, i.e., the position of the object, in a first cycle; and a second identification unit that repeatedly outputs the result of processing for identifying the speed of the object, i.e., the speed of the object, in a second cycle shorter than the first cycle.

[0012] (2): Based on the above (1) scheme, the second identification unit has the function of temporarily identifying the position of an object that is not output by the first identification unit. The identification device also has a fusion unit, which performs the following processing: when the position is output from the first identification unit and the speed is output from the second identification unit, the position output by the first identification unit and the speed output by the second identification unit are output as the state of the object. For objects that are not output by the first identification unit but are output by the second identification unit, the position and speed output by the second identification unit are output as the state of the object.

[0013] (3): Based on the above (2) scheme, the second identification unit has the function of identifying the existence of an object whose position is not output by the first identification unit.

[0014] (4): Based on the above (1) scheme, the first identification unit and the second identification unit perform processing by executing at least some common processing steps respectively.

[0015] (5): Based on the above (1) scheme, the first identification unit and the second identification unit are respectively implemented by one or more processors performing time-sharing processing as the first identification unit and processing as the second identification unit.

[0016] (6): Based on the above (1) scheme, the first identification unit and the second identification unit are respectively processed by separate processors.

[0017] (7): Based on the above (1) scheme, the first identification unit has the function of repeatedly outputting the position and speed of the object in the first cycle, the second identification unit has the function of repeatedly outputting the position and speed of the object in the second cycle, and the identification device also has a fusion unit, which outputs the position output by the first identification unit and the speed output by the second identification unit as the state of the object when the first identification unit outputs for the first time.

[0018] (8): Based on the above (7) scheme, when the fusion unit outputs the position and speed output by the first identification unit as the state of the object after the first identification unit performs the second output.

[0019] (9): Based on the above (1) scheme, the first identification unit has the function of repeatedly outputting the position and speed of the object in the first cycle. The identification device also has a fusion unit, which outputs the position and speed output by the first identification unit as the state of the object when the first identification unit outputs.

[0020] (10): Based on the above (1) scheme, the first identification unit has the function of repeatedly outputting the position and speed of the object in the first cycle, and the second identification unit has the function of repeatedly outputting the position and speed of the object in the second cycle. The identification device also has a fusion unit, which outputs the position output by the first identification unit and the speed output by the first identification unit based on the position output by the first identification unit and the position output by the second identification unit in the previous time before the first identification unit outputs the first identification unit as the state of the object when the first identification unit outputs the first identification unit for the first time.

[0021] (11): Based on any of the above schemes (1) to (10), after the first identification unit outputs the first position at a first time, the first identification unit does not output and the second identification unit outputs at a second time, the second identification unit performs linear interpolation based on the historical information of the position and speed of the object previously identified, thereby updating the first position.

[0022] (12): Another aspect of the present invention relates to a control device for a mobile body, wherein the control device for the mobile body includes: an identification device of the above-described (1) aspect; and a driving control unit that moves the mobile body in a manner that avoids the approach of an object that has output a status to the identification device.

[0023] (13): Another aspect of the present invention relates to an identification method, which is an identification method performed using an identification device, the identification device identifying the position and speed of an object existing in the vicinity of the moving body based on the output of a detection device for detecting the surrounding conditions of the moving body, wherein the identification method includes the following processing: repeatedly outputting the result of processing for identifying the position of the object, i.e., the position of the object, for a first cycle; and repeatedly outputting the result of processing for identifying the speed of the object, i.e., the speed of the object, for a second cycle shorter than the first cycle.

[0024] (14): Another aspect of the present invention relates to a program that causes the processor of an identification device to perform processing, the identification device identifying the position and speed of an object present in the vicinity of the moving body based on the output of a detection device for detecting the surrounding conditions of the moving body, wherein the program is used to cause the processor of the identification device to perform the following processing: repeatedly outputting the result of processing for identifying the position of the object, i.e., the position of the object, for a first cycle; and repeatedly outputting the result of processing for identifying the speed of the object, i.e., the speed of the object, for a second cycle shorter than the first cycle.

[0025] Invention Effects

[0026] According to the schemes (1) to (14), it is possible to quickly follow the changes in the speed of an object. Attached Figure Description

[0027] Figure 1 This is a structural diagram of a vehicle system 1 that utilizes the identification device and the control device for the moving body according to the first embodiment.

[0028] Figure 2 This is a functional structure diagram of the first control unit and the second control unit.

[0029] Figure 3 This is a diagram illustrating an example of the structure of the identification unit.

[0030] Figure 4 This is a diagram illustrating an example of the relationship between the first recognition unit and the second recognition unit.

[0031] Figure 5 This is a diagram illustrating an example of the actions of the first and second recognition units in a certain scene.

[0032] Figure 6 This diagram illustrates the processing of the future location prediction / risk setting department.

[0033] Figure 7 This diagram illustrates an example of the operation of the first and second recognition units in a scenario according to the second embodiment.

[0034] Figure 8 This diagram illustrates an example of the operation of the first and second recognition units in a scenario according to the third embodiment.

[0035] Figure 9 This diagram illustrates an example of the operation of the first and second recognition units in a scenario according to the fourth embodiment. Detailed Implementation

[0036] [summary]

[0037] Hereinafter, embodiments of the identification device, control device for a mobile body, identification method, and procedure of the present invention will be described with reference to the accompanying drawings. The identification device identifies at least the position and speed of objects present around the mobile body. The identification device may be mounted on the mobile body, but it may also be disposed outside the mobile body. The control device for the mobile body controls the drive device of the mobile body to move the mobile body. A mobile body refers to any mobile body capable of moving in space using mechanical power, such as four-wheeled or two-wheeled vehicles, micro-mobile bodies, or drones, aircraft, and ships. A mobile body may be a mobile body carrying people or animals, or it may be a mobile body without human intervention. In the following description, the mobile body refers to a vehicle carrying people and moving on a road; an "autonomous driving control device" will be used as an example of a control device for explanation.

[0038] <First Implementation>

[0039] [Overall Structure]

[0040] Figure 1 This is a structural diagram of a vehicle system 1 that utilizes the identification device and the control device for the moving body according to the first embodiment. The vehicle equipped with vehicle system 1 is, for example, a two-wheeled, three-wheeled, or four-wheeled vehicle, and its drive source is an internal combustion engine such as a diesel engine or a gasoline engine, an electric motor, or a combination thereof. The electric motor operates using electricity generated by a generator connected to the internal combustion engine, or electricity discharged from a secondary battery or fuel cell.

[0041] Vehicle system 1 includes, for example, a camera 10, a radar device 12, a LiDAR (Light Detection and Ranging) system 14, an object recognition device 16, a communication device 20, an HMI (Human Machine Interface) 30, vehicle sensors 40, a navigation device 50, an MPU (Map Positioning Unit) 60, driving controls 80, an automatic driving control device 100, a driving force output device 200, a braking device 210, and a steering device 220. These devices and equipment are interconnected through multiple communication lines such as CAN (Controller Area Network) communication lines, serial communication lines, and wireless communication networks. It should be noted that... Figure 1 The structure shown is just one example; you can omit part of the structure or add other structures.

[0042] Camera 10 is, for example, a digital camera utilizing a solid-state imaging element such as CCD (Charge Coupled Device) or CMOS (Complementary Metal-Oxide Semiconductor). Camera 10 is mounted anywhere on the vehicle equipped with vehicle system 1 (hereinafter referred to as vehicle M). When taking pictures of the front, camera 10 is mounted on the upper part of the windshield, the back of the rearview mirror inside the vehicle, etc. Camera 10, for example, periodically and repeatedly takes pictures of the surroundings of vehicle M. Camera 10 can also be a stereo camera.

[0043] Radar device 12 radiates millimeter-wave and other radio waves around the vehicle M and detects the radio waves reflected by objects (reflected waves) to detect at least the position (distance and orientation) of the objects. Radar device 12 can be installed at any part of the vehicle M. Radar device 12 can also detect the position and speed of objects using FM-CW (Frequency Modulated Continuous Wave) method.

[0044] The LIDAR14 illuminates the periphery of vehicle M with light (or electromagnetic waves of a wavelength close to light) and measures the scattered light. The LIDAR14 determines the distance to the object based on the time from the emission of light to the reception of light. The illuminating light can be, for example, a pulsed laser. The LIDAR14 can be mounted at any location on vehicle M.

[0045] The object recognition device 16 performs sensor fusion processing on the detection results from some or all of the cameras 10, radar device 12, and LIDAR 14 to identify the position, type, speed, etc. of objects. The object recognition device 16 outputs the recognition results to the autonomous driving control device 100. Alternatively, the object recognition device 16 can directly output the detection results from the cameras 10, radar device 12, and LIDAR 14 to the autonomous driving control device 100. The object recognition device 16 can also be omitted from the vehicle system 1.

[0046] The communication device 20 uses, for example, cellular networks, Wi-Fi networks, Bluetooth (registered trademark), DSRC (Dedicated Short Range Communication), etc., to communicate with other vehicles in the vicinity of vehicle M, or communicates with various server devices via wireless base stations.

[0047] The HMI30 provides various information to the occupants of vehicle M and accepts input operations performed by the occupants. The HMI30 includes various display devices, speakers, buzzers, touch panels, switches, buttons, etc.

[0048] The vehicle sensor 40 includes a vehicle speed sensor for detecting the speed of the vehicle M, an acceleration sensor for detecting acceleration, a yaw rate sensor for detecting angular velocity about a vertical axis, and an orientation sensor for detecting the orientation of the vehicle M.

[0049] The navigation device 50 includes, for example, a GNSS (Global Navigation Satellite System) receiver 51, a navigation HMI 52, and a route determination unit 53. The navigation device 50 stores first map information 54 in a storage device such as an HDD (Hard Disk Drive) or flash memory. The GNSS receiver 51 determines the position of the vehicle M based on signals received from GNSS satellites. The position of the vehicle M can also be determined or supplemented by an INS (Inertial Navigation System) utilizing the output of the vehicle sensor 40. The navigation HMI 52 includes a display device, a speaker, a touch panel, buttons, etc. The navigation HMI 52 can also be partially or entirely common to the aforementioned HMI 30. The route determination unit 53, for example, refers to the first map information 54 to determine the route (hereinafter referred to as the map path) from the position of the vehicle M determined by the GNSS receiver 51 (or any input position) to the destination input by the occupant using the navigation HMI 52. The first map information 54 is, for example, information representing the shape of a road by indicating road segments and nodes connecting the road segments. The first map information 54 may also include road curvature, POI (Point of Interest) information, etc. The path on the map is output to the MPU 60. The navigation device 50 can also provide route guidance using the navigation HMI 52 based on the path on the map. The navigation device 50 can also be implemented, for example, through the functions of a terminal device such as a smartphone or tablet held by the occupant. The navigation device 50 can also send its current location and destination to the navigation server via the communication device 20, and obtain the path equivalent to the path on the map from the navigation server.

[0050] MPU 60 includes, for example, a lane recommendation unit 61, and stores the second map information 62 in a storage device such as an HDD or flash memory. The lane recommendation unit 61 divides the path on the map provided by the navigation device 50 into multiple blocks (e.g., every 100 [m] in relation to the vehicle's direction of travel), and determines a recommended lane for each block with reference to the second map information 62. The lane recommendation unit 61 determines which lane to drive in from the left. When the path on the map has branching points, the lane recommendation unit 61 determines the recommended lane in a way that allows the vehicle M to travel on a reasonable path to the branch destination.

[0051] The second map information 62 is map information with higher precision than the first map information 54. The second map information 62 may include, for example, information about the center of a lane or the boundaries of a lane. Furthermore, the second map information 62 may include road information, traffic restriction information, address information (address and postal code), facility information, telephone number information, and information about prohibited areas in Mode A or Mode B (described later). The second map information 62 can be updated in real time by communicating with other devices via the communication device 20.

[0052] The driving control unit 80 includes, for example, a steering wheel, accelerator pedal, brake pedal, gear shift lever, and other control components. Sensors are installed on the driving control unit 80 to detect the amount or presence of operation, and the detection results are output to some or all of the following: the automatic driving control unit 100, the driving force output device 200, the braking device 210, and the steering device 220.

[0053] The autonomous driving control device 100 includes, for example, a first control unit 120 and a second control unit 160. The first control unit 120 and the second control unit 160 are implemented by executing programs (software) using hardware processors such as CPUs (Central Processing Units). Furthermore, some or all of these components can be implemented using hardware (including circuitry) such as LSIs (Large Scale Integration), ASICs (Application Specific Integrated Circuits), FPGAs (Field-Programmable Gate Arrays), and GPUs (Graphics Processing Units), or through the coordinated use of software and hardware. The program can be pre-stored in a storage device such as an HDD or flash memory (a storage device with a non-transitory storage medium) of the autonomous driving control device 100, or stored in a removable storage medium such as a DVD or CD-ROM, and installed in the HDD or flash memory of the autonomous driving control device 100 by mounting the storage medium (a non-transitory storage medium) onto the drive unit.

[0054] The automatic driving control device 100 is an example of a "control device for a moving body", the identification unit 130 is an example of an "identification device", and the action plan generation unit 140 and the second control unit 160 together are an example of a "driving control unit".

[0055] Figure 2This is a functional structure diagram of the first control unit 120 and the second control unit 160. The first control unit 120 includes, for example, a recognition unit 130 and an action plan generation unit 140. The first control unit 120 can, for example, implement AI (Artificial Intelligence) based functions and functions based on pre-given models in parallel. For example, the function of "recognizing intersections" can be achieved by "parallel execution of intersection recognition based on deep learning, etc., and recognition based on pre-given conditions (the existence of signals capable of pattern matching, road signs, etc.), and comprehensively evaluating both sides by scoring them." This ensures the reliability of autonomous driving.

[0056] The recognition unit 130 identifies the presence, position, speed, and acceleration of objects surrounding the vehicle M based on information input from the camera 10, radar device 12, and LIDAR 14 via the object recognition device 16. The position of an object is identified, for example, as its position on absolute coordinates with a representative point of the vehicle M (center of gravity, drive shaft center, etc.) as the origin, and is used for control. The position of an object can be represented by representative points such as its center of gravity or corners, or by a region. The "state" of an object can also include its acceleration, jerk, or "action state" (e.g., whether it is currently or is about to change lanes).

[0057] Additionally, the recognition unit 130 identifies, for example, the lane in which vehicle M is traveling. For instance, the recognition unit 130 compares the pattern of road markings (e.g., an arrangement of solid and dashed lines) obtained from the second map information 62 with the pattern of road markings surrounding vehicle M identified from the image captured by camera 10, thereby identifying the driving lane. It should be noted that the recognition unit 130 is not limited to recognizing road markings; it can also recognize road markings, including road shoulders, curbs, median strips, guardrails, etc., to identify the driving lane. In this recognition, the position of vehicle M obtained from navigation device 50 and the processing results from INS can also be incorporated. Furthermore, the recognition unit 130 identifies temporary stop lines, obstacles, red lights, toll booths, and other road phenomena.

[0058] When identifying a driving lane, the identification unit 130 identifies the position and posture of the vehicle M relative to the driving lane. For example, the identification unit 130 may identify the deviation of the vehicle M's reference point from the center of the lane, and the angle formed by the vehicle M's direction of travel relative to the line connecting the centers of the lanes, as the relative position and posture of the vehicle M relative to the driving lane. Alternatively, the identification unit 130 may identify the position of the vehicle M's reference point relative to any side end (road dividing line or road boundary) of the driving lane as the relative position of the vehicle M relative to the driving lane.

[0059] The action plan generation unit 140 generates a target trajectory for the vehicle M to automatically (without driver intervention) in the future, in a manner that, in principle, the vehicle travels within the recommended lane determined by the recommended lane determination unit 61 and avoids approaching objects identified by the identification unit 130 (excluding objects that can be crossed, such as road markings, road signs, and manholes). For example, the identification unit 130 sets a risk zone centered on the object whose status has been output, and within the risk zone, the identification unit 130 sets the risk as an index value indicating the degree to which the vehicle M should not approach. The action plan generation unit 140 generates the target trajectory in a manner that avoids passing through locations where the risk exceeds a predetermined value. Objects include moving objects, therefore the risk distribution is not set per control cycle, but rather considers the predicted future position of the object based on its speed and sets it for multiple future time points. The target trajectory, for example, includes a speed element. For example, the target trajectory is represented as a trajectory obtained by sequentially arranging the locations (track points) that the vehicle M should reach. Track points are locations that vehicle M should arrive at at predetermined distances (e.g., a few meters) along the route. In contrast, target velocity and target acceleration are generated as part of the target track at predetermined sampling times (e.g., a fraction of a second). Alternatively, track points can also be locations that vehicle M should arrive at at the sampling time at predetermined sampling times. In this case, the target velocity and target acceleration information are represented by the intervals of the track points.

[0060] The second control unit 160 controls the driving force output device 200, the braking device 210 and the steering device 220 so that the vehicle M passes through the target track generated by the action plan generation unit 140 at a predetermined time.

[0061] The second control unit 160 includes, for example, an acquisition unit 162, a speed control unit 164, and a steering control unit 166. The acquisition unit 162 acquires information about the target track (track point) generated by the action plan generation unit 140 and stores this information in a memory (not shown). The speed control unit 164 controls the driving force output device 200 or the braking device 210 based on the speed elements associated with the target track stored in the memory. The steering control unit 166 controls the steering device 220 based on the curvature of the target track stored in the memory. The processing of the speed control unit 164 and the steering control unit 166 can be achieved, for example, through a combination of feedforward control and feedback control. As an example, the steering control unit 166 combines feedforward control corresponding to the curvature of the road ahead of the vehicle M with feedback control based on deviation from the target track.

[0062] The driving force output device 200 outputs driving force (torque) for vehicle movement to the drive wheels. The driving force output device 200 includes, for example, a combination of an internal combustion engine, an electric motor, and a transmission, as well as an ECU (Electronic Control Unit) that controls them. The ECU controls the above-mentioned structure according to information input from the second control unit 160 or from the driving operation device 80.

[0063] The braking device 210 includes, for example, a brake caliper, a hydraulic cylinder that transmits hydraulic pressure to the brake caliper, an electric motor that generates hydraulic pressure in the hydraulic cylinder, and a braking ECU. The braking ECU controls the electric motor according to information input from the second control unit 160 or from the driving operation unit 80, so that braking torque corresponding to the braking operation is output to each wheel. The braking device 210 may have a backup mechanism for transmitting hydraulic pressure generated by the operation of the brake pedal included in the driving operation unit 80 via the master hydraulic cylinder to the hydraulic cylinder. It should be noted that the braking device 210 is not limited to the structure described above, and may also be an electronically controlled hydraulic braking device that controls the actuator according to information input from the second control unit 160, thereby transmitting hydraulic pressure from the master hydraulic cylinder to the hydraulic cylinder.

[0064] The steering system 220 includes, for example, a steering ECU and an electric motor. The electric motor, for example, applies force to a rack and pinion mechanism to change the direction of the steering wheels. The steering ECU drives the electric motor according to information input from the second control unit 160 or from the driving control unit 80, thereby changing the direction of the steering wheels.

[0065] [More detailed structure of the identification unit]

[0066] Here, a more detailed description of the structure of the identification unit 130 will be provided. In order to achieve the above-described functions, the identification unit 130 has the structure described below. Figure 3 This is a diagram showing an example of the structure of the identification unit 130. The identification unit 130 includes, for example, a first identification unit 132, a second identification unit 134, a fusion unit 136, and a future location prediction / risk setting unit 138.

[0067] The first recognition unit 132 repeatedly outputs the result of the processing used to recognize the position of an object in a first cycle, namely the position of the object.

[0068] The second recognition unit 134 repeatedly outputs the result of processing for recognizing the speed of an object—that is, the speed of the object—at a second cycle shorter than the first cycle. In other words, the second recognition unit 134 processes at a higher speed than the first recognition unit 132. For example, the second recognition unit 134 processes at a frequency several to ten times that of the first recognition unit 132. The second recognition unit 134 has the function of temporarily recognizing the presence and position of objects whose positions are not output by the first recognition unit 132. There are no special limitations on this function; it can be a simpler and less demanding process compared to the processing performed by the first recognition unit 132. As an example, this function can be achieved by combining contour extraction processing with size recognition processing. This function can stop when the first recognition unit 132 begins outputting a position, or it can continue after the first recognition unit 132 begins outputting a position.

[0069] When the fusion unit 136 outputs the position from the first recognition unit 132 and the velocity (and position) from the second recognition unit 134, it outputs the position output by the first recognition unit 132 and the velocity output by the second recognition unit 134 as the state of the object. For objects for which the position is not output from the first recognition unit 132 but the position (the position temporarily recognized as described above) and velocity are output from the second recognition unit 134, the position and velocity output by the second recognition unit 134 are output as the state of the object.

[0070] The first recognition unit 132 and the second recognition unit 134 perform processing by executing at least some common processing steps. For example, the first recognition unit 132 and the second recognition unit 134 respectively identify the position, speed, etc. of objects by inputting images captured by the camera 10 (or images obtained by preprocessing them) into a model that has been trained by a DNN (Deep Neural Network) or the like, and output them. The second recognition unit 134 compresses the input image to a lower resolution than the image input to the first recognition unit 132, thereby processing the image at a higher speed than the first recognition unit 132. Figure 4 This diagram illustrates an example of the relationship between the first recognition unit 132 and the second recognition unit 134. It should be noted that in this case, the input image input to the first recognition unit 132 is time-staggered, resulting in fewer frames compared to the input image input to the second recognition unit 134.

[0071] Figure 5This diagram illustrates an example of the operation of the first recognition unit 132 and the second recognition unit 134 in a certain scene. In this diagram, the second recognition unit 134 processes data at a frequency three times that of the first recognition unit 132. The control timing is an imaginary moment that occurs every predetermined period (the same as the processing period of the second recognition unit 134 in this embodiment). At each control timing, the image captured by the camera 10 is repeatedly input. In this diagram, it is provisionally set that the control timing starts from 1.

[0072] At control time 1, the second identification unit 134 initially and tentatively identifies the presence and position of the object. The second identification unit 134 also identifies the object's velocity. At this time, the first identification unit 132 or the second identification unit 134 (or other functional units) calculates, for example, the covariance matrix generated by arranging the states calculated previously for this control time, its intrinsic values, and the interpolation values ​​used for future expansion, thereby predicting (predicting the next) the object's position and velocity for the next control time 2. This prediction process is repeated for each control time.

[0073] At control moment 1, regarding the definition of the state output by the recognition unit 130, the object's position is defined as tentatively recognized, and its speed is defined as deterministically recognized. The degree of control for that object is adjusted based on whether it is tentatively or deterministically recognized. For example, for an object whose position is tentatively recognized, the risk is calculated to be smaller compared to an object whose position is deterministically recognized, and thus the degree of control is relaxed.

[0074] During control timing 2, the second recognition unit 134 updates the position and velocity of the object based on the prediction result of control timing 1 and the result of processing the input image.

[0075] At control time 3, the first recognition unit 132 updates the object's position based on the prediction result of control time 2 and the result of processing the input image. Meanwhile, the second recognition unit 134 updates the object's velocity based on the prediction result of control time 2 and the result of processing the input image. At this point, the existence of the object is confirmed, and the object's position is defined as being deterministically recognized.

[0076] During control timing 4, the second recognition unit 134 updates the object's speed based on the prediction result of control timing 3 and the result of processing the input image.

[0077] At control timing 5, the second recognition unit 134 updates the object's speed based on the prediction result of control timing 4 and the result of processing the input image.

[0078] At control time 6, the first recognition unit 132 updates the object's position based on the prediction result of control time 5 and the result of processing the input image. Meanwhile, the second recognition unit 134 updates the object's velocity based on the prediction result of control time 5 and the result of processing the input image. Thereafter, the same processing as control times 4-6 is repeatedly executed.

[0079] The future position prediction / risk setting unit 138 predicts the future position of the object based on the object's state (position, velocity) output by the fusion unit 136, and sets a risk for the object. Figure 6 This diagram illustrates the processing of the future position prediction / risk setting unit 138. The future position prediction / risk setting unit 138 sets an index value, or risk, representing the degree to which the vehicle M should not enter or approach in an imaginary plane, i.e., a hypothetical plane S, which is a two-dimensional plane representing the space surrounding the vehicle M when viewed from above. Regarding risk, the higher the risk value, the less the vehicle M should enter or approach; the closer the risk value is to zero, the more preferable the vehicle M's movement. However, this relationship can also be reversed. In the case where the moving body is not a vehicle but a flying object such as a drone, the future position prediction / risk setting unit 138 can perform the same processing in three-dimensional space instead of the hypothetical plane S.

[0080] The future position prediction / risk setting unit 138 sets the risk in the hypothetical plane S not only about the current time point, but also about the future position of the object predicted in advance by a certain time interval, such as the current time point, Δt after (time t+Δt), 2Δt after (time t+2Δt), ...

[0081] The future position prediction / risk setting unit 138 sets risks on a hypothetical plane S using ellipses or circles as contour lines based on the direction and speed of travel for traffic participants (moving objects), such as vehicles, pedestrians, and bicycles. It sets fixed risk values ​​for drivable areas. In the diagram, DM represents the direction of travel of vehicle M. R(M1) represents the risk of the stopped vehicle M1, and R(P) represents the risk of pedestrian P. Pedestrian P moves along the direction of crossing the road, so risks are set at different locations at future points in time, different from the current time. The same applies to moving vehicles, bicycles, etc. R(BD) represents the risk of the drivable area BD. In the diagram, the density of the shading line represents the risk value; the denser the shading line, the greater the risk. The future position prediction / risk setting unit 138 can also set risks in a manner where the value increases the further away from the center of the lane.

[0082] By processing data in this structure, the recognition unit 130 and the automatic driving control device 100 can quickly follow changes in the speed of an object. For example, in Figures 3-5As explained, the second recognition unit 134 processes the input image using compressed data, thus shortening the overall processing time and enabling faster processing compared to the first recognition unit 132, which uses high-resolution images. The advantage of processing with high-resolution images is primarily its ability to accurately detect objects located far away from the vehicle M, but it doesn't show a corresponding performance difference for objects near the vehicle M. On the other hand, when performing risk assessment based on the future position of an object, it's desirable to quickly follow sudden changes in the object's speed (e.g., a pedestrian suddenly darting out, another vehicle abruptly stopping). In this regard, the second recognition unit 134, which primarily recognizes speed, in the recognition unit 130 of this embodiment is configured for high-speed processing, enabling rapid detection of sudden changes in the object's speed. Therefore, it's possible to quickly follow changes in the object's speed and, consequently, rapidly control the behavior of the vehicle M.

[0083] The first identification unit 132 and the second identification unit 134 are implemented, for example, by time-sharing processing of the first identification unit 132 and the second identification unit 134 by one or more processors. The same applies to the fusion unit 136 and the future location prediction / risk setting unit 138. Alternatively, the first identification unit 132 and the second identification unit 134 can be implemented by separate processors. The same applies to the fusion unit 136 and the future location prediction / risk setting unit 138.

[0084] <Second Implementation>

[0085] Next, the second embodiment will be described. The second embodiment differs from the first embodiment in that the first recognition unit 132 starts outputting the position at the point when it successfully recognizes the object's position, and outputs both the position and velocity after recognizing the object's velocity. Furthermore, the second embodiment differs from the first embodiment in that, after the first recognition unit 132 recognizes the object's position, during control periods when the first recognition unit 132 does not perform any action, the object's position is updated based on the velocity recognized by the second recognition unit 134. In the following description, structures having the same functions as those in the first embodiment will be labeled with the same reference numerals and names, and their specific descriptions will be omitted.

[0086] Figure 7 This diagram illustrates an example of the operation of the first recognition unit 132 and the second recognition unit 134 in a scenario according to the second embodiment. In this diagram, the second recognition unit 134 processes data at a frequency three times that of the first recognition unit 132.

[0087] At control time 1, the second recognition unit 134 initially and tentatively recognizes the presence and position of the object. The second recognition unit 134 also recognizes the object's velocity. That is, the position and velocity are recognized based on the object recognition result ("high-speed object recognition result") of the second recognition unit 134, which performs object recognition at high speed. At this time, the second recognition unit 134 (or other functional units) calculates, for example, the covariance matrix generated by arranging the states calculated previously for this control time, its intrinsic values, and the interpolation values ​​used for future expansion, thereby predicting (predicting the next) the position and velocity of the object at the next control time 2. This prediction process is repeated for each control time. At control time 1, regarding the definition of the state output by the recognition unit 130, the object's position is defined as tentatively recognized, and its velocity is defined as deterministically recognized.

[0088] During control timing 2, the second recognition unit 134 updates the position and velocity of the object based on the prediction result of control timing 1 and the result of processing the input image.

[0089] Control timing 3 is the timing when the first recognition unit 132 first outputs. At control timing 3, the first recognition unit 132 updates the object's position based on the prediction result from control timing 2 and the result of processing the input image. That is, the object's position is updated based on the object recognition result ("normal object recognition result") identified by the first recognition unit 132 performing object recognition at a normal speed. Furthermore, the second recognition unit 134 updates the object's speed based on the prediction result from control timing 2 and the result of processing the input image. That is, the object's speed is updated based on the high-speed object recognition result identified by the second recognition unit 134. At this time, the existence of the object is determined, and the object's position is defined as being deterministically identified.

[0090] At control time 4, the second recognition unit 134 updates the object's velocity based on the prediction result of control time 3 and the result of processing the input image. Furthermore, the second recognition unit 134 updates the object's position based on the velocity. For example, the second recognition unit 134 updates the object's position based on linear interpolation. The second recognition unit 134 uses the object's position identified by the first recognition unit 132 at control time 3 as a reference position, adds the object's movement amount (movement amount from control time 3 to control time 4) calculated using linear interpolation to this reference position, thereby estimating the current object's position and updating the object's position with the estimated position. In linear interpolation, for example, a covariance matrix (history information of position and velocity) generated by arranging the states calculated previously at the current control time, and its intrinsic values, are used.

[0091] At control time 5, the second recognition unit 134 updates the object's velocity based on the prediction result of control time 4 and the result of processing the input image. Furthermore, the second recognition unit 134 updates the object's position based on the velocity. For example, the second recognition unit 134 updates the object's position using linear interpolation. The second recognition unit 134 uses the object's position updated at control time 3 as a reference position, adds the object's movement amount (movement amount from control time 3 to control time 5) calculated using linear interpolation to this reference position, thereby estimating the current object's position, and updates the object's position based on the estimated position. Alternatively, the second recognition unit 134 uses the object's position updated at control time 4 as a reference position, adds the object's movement amount (movement amount from control time 4 to control time 5) calculated using linear interpolation to this reference position, thereby estimating the current object's position, and updates the object's position based on the estimated position.

[0092] That is, after the first identification unit 132 outputs the first position at the first time, the first identification unit 132 does not output and the second identification unit 134 outputs at the second time. The second identification unit 134 performs linear interpolation based on the historical information of the position and velocity of the previously identified object, thereby updating the first position.

[0093] Control timing 6 is the timing when the first recognition unit 132 outputs its second output. At control timing 6, the first recognition unit 132 updates the object's position based on the prediction result of control timing 5 and the result of processing the input image. Furthermore, at control timing 6, information about the object's position recognized by the first recognition unit 132 is accumulated, enabling velocity estimation based on this position information. Therefore, the first recognition unit 132 updates the object's velocity based on the recognition result of the previously recognized object's position (e.g., the recognition result of control timing 3) and the result of processing the input image. Thereafter, the same processing as control timings 4-6 is repeatedly executed.

[0094] That is, the first recognition unit 132 has the function of repeatedly outputting the position and velocity of an object in a first cycle. The second recognition unit 134 has the function of repeatedly outputting the position and velocity of an object in a second cycle. When the first recognition unit 132 outputs for the first time, the fusion unit 136 outputs the position output by the first recognition unit 132 and the velocity output by the second recognition unit 134 as the state of the object. In addition, when the first recognition unit 132 outputs for the second time or subsequent times, the fusion unit 136 outputs the position and velocity output by the first recognition unit 132 as the state of the object.

[0095] According to the second embodiment described above, it is possible to quickly follow changes in the speed of an object, thereby enabling rapid control of the behavior of the moving body. Furthermore, by outputting the position and speed detected by the first recognition unit 132 as the object's state after the object's speed is detected, the accuracy of the object's position and speed estimation can be improved. Moreover, since the first recognition unit 132 does not perform any control actions after detecting the object's position, the second recognition unit 134 updates the object's position based on the speed, thereby improving the accuracy of the object's position estimation and effectively reducing the impact of position errors on the fusion unit, thus improving fusion performance.

[0096] <Third Implementation Method>

[0097] Next, the third embodiment will be described. The third embodiment differs from the first and second embodiments in that the first identification unit 132 does not output anything before identifying the speed, and outputs both the speed and position after identifying the speed. In the following description, structures having the same functions as those in the first and second embodiments will be labeled with the same reference numerals and names, and their specific descriptions will be omitted.

[0098] Figure 8 This diagram illustrates an example of the operation of the first recognition unit 132 and the second recognition unit 134 in a scenario according to the third embodiment. In this diagram, the second recognition unit 134 processes data at a frequency three times that of the first recognition unit 132.

[0099] At control time 1, the second identification unit 134 initially and tentatively identifies the presence and position of the object. The second identification unit 134 also identifies the object's velocity. At this time, the second identification unit 134 (or other functional units) calculates, for example, the covariance matrix generated from the arrangement of states calculated before this control time, its intrinsic values, and the interpolation values ​​used for future expansion, thereby predicting (predicting the next) the object's position and velocity at the next control time 2. This prediction process is repeated for each control time. At control time 1, regarding the definition of the state output by the identification unit 130, the object's position is defined as tentatively identified, and its velocity is defined as deterministically identified.

[0100] During control timing 2, the second recognition unit 134 updates the position and velocity of the object based on the prediction result of control timing 1 and the result of processing the input image.

[0101] At control time 3, the second recognition unit 134 updates the object's position and velocity based on the prediction result from control time 2 and the result of processing the input image. Furthermore, control time 3 is the time when the first recognition unit 132 first performs its operation. At this control time 3, the first recognition unit 132 identifies the object's position based on the prediction result from control time 2 and the result of processing the input image, but does not output the identified position; instead, it stores it in memory (not shown).

[0102] During control timing 4, the second recognition unit 134 updates the position and velocity of the object based on the prediction result of control timing 3 and the result of processing the input image.

[0103] During control timing 5, the second recognition unit 134 updates the position and velocity of the object based on the prediction result of control timing 4 and the result of processing the input image.

[0104] Control timing 6 is the timing when the first recognition unit 132 first outputs. At control timing 6, the first recognition unit 132 updates the object's position based on the prediction result from control timing 5 and the result of processing the input image. At this time, the existence of the object is confirmed, and the object's position is defined as being deterministically identified. Furthermore, at control timing 6, information about the object's position identified by the first recognition unit 132 is accumulated, making it possible to perform a velocity estimation based on this position information. Therefore, the first recognition unit 132 updates the object's velocity based on the recognition result of previously identified object positions (e.g., the recognition result from control timing 3) and the result of processing the input image.

[0105] At control time 7, the second recognition unit 134 updates the object's velocity based on the prediction result of control time 6 and the result of processing the input image. Furthermore, the second recognition unit 134 updates the object's position based on the velocity. For example, the second recognition unit 134 updates the object's position using linear interpolation. The second recognition unit 134 uses the object's position updated at control time 6 as a reference position, adds the object's movement amount calculated using linear interpolation (the movement amount from control time 6 to control time 7) to this reference position, thereby estimating the current object position, and updates the object's position based on the estimated position.

[0106] At control time 8, the second recognition unit 134 updates the object's velocity based on the prediction result of control time 7 and the result of processing the input image. Furthermore, the second recognition unit 134 updates the object's position based on the velocity. For example, the second recognition unit 134 updates the object's position using linear interpolation. The second recognition unit 134 uses the object's position updated at control time 6 as a reference position, adds the object's movement amount (movement from control time 6 to control time 8) calculated using linear interpolation to this reference position, thereby estimating the current object's position, and updates the object's position based on the estimated position. Alternatively, the second recognition unit 134 uses the object's position updated at control time 7 as a reference position, adds the object's movement amount (movement from control time 7 to control time 8) calculated using linear interpolation to this reference position, thereby estimating the current object's position, and updates the object's position based on the estimated position.

[0107] Control timing 9 is the timing when the first recognition unit 132 performs its second output. At control timing 9, the first recognition unit 132 updates the object's position based on the prediction result of control timing 8 and the result of processing the input image. Additionally, the first recognition unit 132 updates the object's velocity based on the recognition results of previously recognized object positions (e.g., the recognition results of control timings 3 and 6) and the result of processing the input image. Thus, at the timing when the first recognition unit 132 outputs, the fusion unit 136 outputs the position and velocity output by the first recognition unit 132 as the object's state. Thereafter, the same processing as control timings 7-9 is repeatedly executed.

[0108] According to the third embodiment described above, it is possible to quickly follow changes in the speed of an object, thereby enabling rapid control of the behavior of the moving body. Furthermore, after accumulating information about the object's position detected by the first recognition unit 132 to estimate its speed, the first recognition unit 132 updates the object's speed based on the speed detected by the first recognition unit 132 when controlling the action. This improves the accuracy of the object's position and speed estimation. Additionally, it maintains design versatility, such as assembling a second recognition unit 134 into existing systems that only have the first recognition unit 132.

[0109] <Fourth Implementation>

[0110] Next, the fourth embodiment will be described. The fourth embodiment differs from the first to third embodiments in that, at the initial control timing of the first recognition unit 132's operation, the object's speed is updated based on the recognition result identified by the second recognition unit 134 under the previous control timing and the recognition result identified by the first recognition unit 132 under the current control timing. In the following description, structures having the same functions as those in the first to third embodiments will be labeled with the same reference numerals and names, and their specific descriptions will be omitted.

[0111] Figure 9 This diagram illustrates an example of the operation of the first recognition unit 132 and the second recognition unit 134 in a scenario according to the fourth embodiment. In this diagram, the second recognition unit 134 processes data at a frequency three times that of the first recognition unit 132.

[0112] At control time 1, the second identification unit 134 initially and tentatively identifies the presence and position of the object. The second identification unit 134 also identifies the object's velocity. At this time, the second identification unit 134 (or other functional units) calculates, for example, the covariance matrix generated by arranging the states calculated previously for this control time, its intrinsic values, and the interpolation values ​​for future expansion, thereby predicting (predicting the next) the object's position and velocity for the next control time 2. This prediction process is repeated for each control time. At control time 1, regarding the definition of the state output by the identification unit 130, the object's position is defined as tentatively identified, and its velocity is defined as deterministically identified.

[0113] During control timing 2, the second recognition unit 134 updates the position and velocity of the object based on the prediction result of control timing 1 and the result of processing the input image.

[0114] Control timing 3 is the timing when the first recognition unit 132 first outputs. At control timing 3, the first recognition unit 132 updates the object's position based on the prediction result of control timing 2 and the result of processing the input image. At this time, the existence of the object is confirmed, and the object's position is defined as being deterministically recognized. Furthermore, the first recognition unit 132 updates the object's velocity based on the recognition result identified by the second recognition unit 134 at control timing 2 (the position of the previous high-speed object recognition result) and the recognition result identified by the first recognition unit 132 at control timing 3 (the position of the current normal object recognition result). Therefore, the fusion unit 136 outputs the position output by the first recognition unit 132 and the velocity output by the first recognition unit 132 based on the position output by the first recognition unit 132 and the position output by the second recognition unit 134 at the previous timing, as the object's state, at the timing when the first recognition unit 132 first outputs.

[0115] At control time 4, the second recognition unit 134 updates the object's velocity based on the prediction result of control time 3 and the result of processing the input image. Furthermore, the second recognition unit 134 updates the object's position based on the velocity. For example, the second recognition unit 134 updates the object's position using linear interpolation. The second recognition unit 134 uses the object's position identified by the first recognition unit 132 at control time 3 as a reference position, adds the object's movement amount (the movement amount from control time 3 to control time 4) calculated using linear interpolation to this reference position, thereby estimating the current object's position, and updates the object's position based on the estimated position.

[0116] At control time 5, the second recognition unit 134 updates the object's velocity based on the prediction result of control time 4 and the result of processing the input image. Furthermore, the second recognition unit 134 updates the object's position based on the velocity. For example, the second recognition unit 134 updates the object's position using linear interpolation. The second recognition unit 134 uses the object's position updated at control time 3 as a reference position, adds the object's movement amount (movement amount from control time 3 to control time 5) calculated using linear interpolation to this reference position, thereby estimating the current object's position, and updates the object's position based on the estimated position. Alternatively, the second recognition unit 134 uses the object's position updated at control time 4 as a reference position, adds the object's movement amount (movement amount from control time 4 to control time 5) calculated using linear interpolation to this reference position, thereby estimating the current object's position, and updates the object's position based on the estimated position.

[0117] Control timing 6 is the timing when the first recognition unit 132 outputs its second output. At control timing 6, the first recognition unit 132 updates the object's position based on the prediction result of control timing 5 and the result of processing the input image. Furthermore, at control timing 6, information about the object's position recognized by the first recognition unit 132 is accumulated, enabling velocity estimation based on this position information. Therefore, the first recognition unit 132 updates the object's velocity based on the recognition result of the previously recognized object's position (e.g., the recognition result of control timing 3) and the result of processing the input image. Thereafter, the same processing as control timings 4-6 is repeatedly executed.

[0118] According to the fourth embodiment described above, it is possible to quickly follow changes in the speed of an object, thereby enabling rapid control of the behavior of the moving object. Furthermore, at the initial control moment when the first recognition unit 132 performs an action, the object's speed is updated based on the recognition result identified by the second recognition unit 134 at the previous moment and the recognition result identified by the first recognition unit 132 at the current control moment. This maintains the design versatility for situations such as assembling a second recognition unit 134 into an existing system that only has a first recognition unit 132.

[0119] It should be noted that, in the above description, the control device is mounted on the vehicle M (i.e., the mobile body), but it is not limited to this. It can also be set in a place separate from the mobile body, and obtain the output data of the camera 10, radar device 12, etc. through communication, and send the drive instruction signal to the mobile body, that is, remotely control the mobile body.

[0120] Furthermore, the embodiments described above are examples, and the present invention is not limited to the structures of these embodiments. The functions or structures included in each embodiment can also be appropriately combined. For example, the following structure can be incorporated into the first embodiment: the control timing where the first identification unit 132 does not operate after the object's position is identified by the first identification unit 132, as described in the second to fourth embodiments, updates the object's position based on the speed identified by the second identification unit 134.

[0121] The implementation methods described above can be performed as follows.

[0122] A control device is a recognition device that identifies the position and velocity of objects present in the vicinity of a moving body based on the output of a detection device for detecting the surrounding conditions of the moving body, wherein...

[0123] The control device includes:

[0124] One or more storage media for storing computer-readable instructions; and

[0125] The processor connected to the one or more storage media,

[0126] The processor performs the following processing by executing computer-readable instructions:

[0127] The result of the processing used to identify the position of the object is repeatedly output in the first cycle, which is the position of the object; and

[0128] The result of the processing used to identify the speed of the object is repeatedly output in a second cycle that is shorter than the first cycle, i.e., the speed of the object.

[0129] The above describes specific embodiments of the present invention, but the present invention is not limited to such embodiments in any way. Various modifications and substitutions can be made without departing from the spirit of the present invention.

[0130] Explanation of reference numerals in the attached figures:

[0131] 10 cameras

[0132] 12 Radar Devices

[0133] 14 LIDAR

[0134] 16 Object recognition devices

[0135] 100 Automatic driving control device

[0136] 120 First Control Department

[0137] 130 Identification Department

[0138] 132 First Identification Department

[0139] 134 Second Identification Department

[0140] 136 Integration Department

[0141] 138 Future Location Prediction / Risk Setting Department

[0142] 140 Action Plan Generation Department

[0143] 160 Second Control Unit. Claims (as amended under Article 19 of the Treaty) 1. [Modified] An identification device that identifies the position and velocity of objects present in the vicinity of a moving body based on the output of a detection device for detecting the surrounding conditions of the moving body, wherein, The identification device includes: A first identification unit repeatedly outputs the result of processing for identifying the position of the object, i.e., the position of the object, in a first cycle; and The second recognition unit repeatedly outputs the result of processing for recognizing the speed of the object, i.e., the speed of the object, at a second cycle shorter than the first cycle. Compared to the processing performed by the first recognition unit to identify the position of the object, the second recognition unit uses a low-resolution image with lower resolution to perform the processing to identify the speed of the object. 2. The identification device according to claim 1, wherein, The second identification unit has the function of temporarily identifying the position of an object whose position has not been output by the first identification unit. The identification device also includes a fusion section, which performs the following processing: When the position is output from the first recognition unit and the velocity is output from the second recognition unit, the position output by the first recognition unit and the velocity output by the second recognition unit are output as the state of the object. For an object that does not output its position from the first recognition unit but outputs its position and velocity from the second recognition unit, the position and velocity output by the second recognition unit are output as the state of the object. 3. The identification device according to claim 2, wherein, The second identification unit has the function of identifying the presence of an object whose position is not output by the first identification unit. 4. The identification device according to claim 1, wherein, The first identification unit and the second identification unit perform processing by executing at least some common processing steps. 5. The identification device according to claim 1, wherein, The first identification unit and the second identification unit are implemented by one or more processors performing time-sharing processing as the first identification unit and processing as the second identification unit, respectively. 6. The identification device according to claim 1, wherein, The first identification unit and the second identification unit are respectively processed by separate processors. 7. The identification device according to claim 1, wherein, The first identification unit has the function of repeatedly outputting the position and velocity of the object in the first cycle. The second recognition unit has the function of repeatedly outputting the position and velocity of the object in the second cycle. The identification device also includes a fusion unit, which outputs the position output by the first identification unit and the speed output by the second identification unit as the state of the object when the first identification unit first outputs. 8. The identification device according to claim 7, wherein, When the fusion unit outputs the position and velocity output by the first identification unit after the first identification unit performs its second output, it outputs the position and velocity output by the first identification unit as the state of the object. 9. The identification device according to claim 1, wherein, The first identification unit has the function of repeatedly outputting the position and velocity of the object in the first cycle. The identification device also includes a fusion unit, which outputs the position and velocity output by the first identification unit as the state of the object when the first identification unit outputs. 10. The identification device according to claim 1, wherein, The first identification unit has the function of repeatedly outputting the position and velocity of the object in the first cycle. The second recognition unit has the function of repeatedly outputting the position and velocity of the object in the second cycle. The identification device further includes a fusion unit that, at the moment when the first identification unit first outputs, outputs the position output by the first identification unit and the speed output by the first identification unit based on the position output by the first identification unit and the position output by the second identification unit at the previous moment before the moment, as the state of the object. 11. The identification device according to any one of claims 1 to 10, wherein, After the first identification unit outputs the first position at a first opportune moment, the first identification unit does not output and the second identification unit outputs at a second opportune moment. The second identification unit performs linear interpolation based on the historical information of the position and velocity of the object that has been previously identified, thereby updating the first position. 12. A control device for a moving body, wherein, The control device for the moving body includes: The identification device according to claim 1; and The driving control unit moves the moving body in a manner that prevents an object that has output a status to the identification device from approaching. 13. [Modified] An identification method, which is an identification method performed using an identification device, said identification device identifying the position and velocity of objects present in the vicinity of the moving body based on the output of a detection device for detecting the surrounding conditions of the moving body, wherein, The identification method includes the following processing: The result of the processing used to identify the position of the object is repeatedly output in the first cycle, which is the position of the object; and The result of the processing used to identify the velocity of the object is repeatedly output in a second cycle that is shorter than the first cycle, i.e., the velocity of the object. Compared to the processing used to identify the position of the object, the processing used to identify the velocity of the object is performed using a low-resolution image with a lower resolution. 14. [Modified] A program that causes a processor of an identification device to perform processing, the identification device identifying the position and velocity of objects present in the vicinity of the moving body based on the output of a detection device for detecting the surrounding conditions of the moving body, wherein, The program is used to cause the processor of the identification device to perform the following processing: The result of the processing used to identify the position of the object is repeatedly output in the first cycle, which is the position of the object; and The result of the processing used to identify the velocity of the object is repeatedly output in a second cycle that is shorter than the first cycle, i.e., the velocity of the object. Compared to the processing used to identify the position of the object, the processing used to identify the velocity of the object is performed using a low-resolution image with a lower resolution. 15. [Additionally] The identification device according to claim 1, wherein, The second identification unit performs a lower-load process compared to the process performed by the first identification unit. 16. [Additionally] The identification device according to claim 1, wherein, The identification device also includes a setting unit that defines a non-driving zone for the object based on its direction of travel or speed.

Claims

1. A recognition device that identifies the position and velocity of an object present in the vicinity of a moving body based on the output of a detection device for detecting the surrounding conditions of the moving body, wherein, The identification device includes: The first identification unit repeatedly outputs the result of processing for identifying the position of the object, i.e., the position of the object, in a first cycle; as well as The second recognition unit repeatedly outputs the result of processing for recognizing the speed of the object, i.e., the speed of the object, in a second cycle shorter than the first cycle.

2. The identification device according to claim 1, wherein, The second identification unit has the function of temporarily identifying the position of an object whose position has not been output by the first identification unit. The identification device also includes a fusion section, which performs the following processing: When the position is output from the first recognition unit and the velocity is output from the second recognition unit, the position output by the first recognition unit and the velocity output by the second recognition unit are output as the state of the object. For an object that does not output its position from the first recognition unit but outputs its position and velocity from the second recognition unit, the position and velocity output by the second recognition unit are output as the state of the object.

3. The identification device according to claim 2, wherein, The second identification unit has the function of identifying the presence of an object whose position is not output by the first identification unit.

4. The identification device according to claim 1, wherein, The first identification unit and the second identification unit perform processing by executing at least some common processing steps.

5. The identification device according to claim 1, wherein, The first identification unit and the second identification unit are implemented by one or more processors performing time-sharing processing as the first identification unit and processing as the second identification unit, respectively.

6. The identification device according to claim 1, wherein, The first identification unit and the second identification unit are respectively processed by separate processors.

7. The identification device according to claim 1, wherein, The first identification unit has the function of repeatedly outputting the position and velocity of the object in the first cycle. The second recognition unit has the function of repeatedly outputting the position and velocity of the object in the second cycle. The identification device also includes a fusion unit, which outputs the position output by the first identification unit and the speed output by the second identification unit as the state of the object when the first identification unit first outputs.

8. The identification device according to claim 7, wherein, When the fusion unit outputs the position and velocity output by the first identification unit after the first identification unit performs its second output, it outputs the position and velocity output by the first identification unit as the state of the object.

9. The identification device according to claim 1, wherein, The first identification unit has the function of repeatedly outputting the position and velocity of the object in the first cycle. The identification device also includes a fusion unit, which outputs the position and velocity output by the first identification unit as the state of the object when the first identification unit outputs.

10. The identification device according to claim 1, wherein, The first identification unit has the function of repeatedly outputting the position and velocity of the object in the first cycle. The second recognition unit has the function of repeatedly outputting the position and velocity of the object in the second cycle. The identification device further includes a fusion unit that, at the moment when the first identification unit first outputs, outputs the position output by the first identification unit and the speed output by the first identification unit based on the position output by the first identification unit and the position output by the second identification unit at the previous moment before the moment, as the state of the object.

11. The identification device according to any one of claims 1 to 10, wherein, After the first identification unit outputs the first position at a first opportune moment, the first identification unit does not output and the second identification unit outputs at a second opportune moment. The second identification unit performs linear interpolation based on the historical information of the position and velocity of the object that has been previously identified, thereby updating the first position.

12. A control device for a moving body, wherein, The control device for the moving body includes: The identification device according to claim 1; and The driving control unit moves the moving body in a manner that prevents an object that has output a status to the identification device from approaching.

13. A recognition method, which is an recognition method performed using a recognition device, said recognition device identifying the position and velocity of objects present in the vicinity of the moving body based on the output of a detection device for detecting the surrounding conditions of the moving body, wherein, The identification method includes the following processing: The result of the processing used to identify the position of the object is repeatedly output in the first cycle, which is the position of the object; and The result of the processing used to identify the speed of the object is repeatedly output in a second cycle that is shorter than the first cycle, i.e., the speed of the object.

14. A program that causes a processor of an identification device to perform processing, the identification device identifying the position and velocity of objects present in the vicinity of the moving body based on the output of a detection device for detecting the surrounding conditions of the moving body, wherein, The program is used to cause the processor of the identification device to perform the following processing: The result of the processing used to identify the position of the object is repeatedly output in the first cycle, which is the position of the object; and The result of the processing used to identify the speed of the object is repeatedly output in a second cycle that is shorter than the first cycle, i.e., the speed of the object.

Images