Information processing device

Abstract

An information processing device according to the present invention comprises a recognition rate estimation unit configured such that when a video captured at a certain point by a vehicle traveling at a certain speed is transmitted to a monitor at a certain bit rate over a communication network, the recognition rate estimation unit estimates, on the basis of the bit rate and the speed, an object recognition rate in the video by the monitor for a location which is apart from the certain point by a stopping distance at the speed. Thus, the safety of a remote automatic driving system is improved.

Description

Information processing device

[0001] The present invention relates to an information processing device.

[0002] Systems that use autonomous driving technology to enable automobiles to drive autonomously are expected to alleviate traffic congestion, improve logistics efficiency, reduce environmental impact, etc., and many studies and demonstration experiments are being conducted. Among these, this article focuses on remote automated driving systems that allow a monitor / operator located remotely from the automobile to operate the vehicle in case of an emergency, etc.

[0003] One of the criteria for conducting demonstration experiments of remotely controlled automated driving systems on public roads is that "even when traveling at the maximum speed specified in the application, the remote monitor / operator must be able to grasp the situation around the experimental vehicle and the direction of travel through video and audio to the same extent as a normal car driver" (Non-Patent Document 1). Therefore, video data must be transmitted from the monitored vehicle to the remote monitor / operator at a video quality that allows the remote monitor / operator to recognize hazards. However, automated driving vehicles must use wireless networks, and the quality of wireless networks is prone to fluctuation in high-speed moving environments, making it impossible to always ensure sufficient communication quality.

[0004] To address this issue of fluctuations in communication quality, a method for controlling the bit rate in accordance with fluctuations in communication quality has been proposed (Non-Patent Document 2). This technology assumes a situation in which an object that obstructs the vehicle's travel path is present, and proposes a video bit rate control technology aimed at achieving a quality that allows the object to be detected at least once within a certain period of time from the moment the object comes into view and becomes recognizable to the moment the autonomous vehicle can stop. Specifically, when the throughput is low, the bit rate is controlled so that the quality exceeds a predetermined threshold value at least once within the certain period, and when the throughput is high, the bit rate is controlled so that the quality is as high as possible while avoiding playback interruptions.

[0005] "Road Use Permit Criteria for Public Road Demonstration Tests of Autonomous Driving," National Police Agency, [online], Internet <URL: https: / / www.npa.go.jp / bureau / traffic / selfdriving / roadtesting / 230403jidountendourosiyoukyokakijyun.pdf> Yokota et al., "Bitrate Control Method for Remote Monitoring Video of Autonomous Driving Vehicles," IEICE Technical Report, vol. 122, no. 184, CQ2022-45, pp. 115-119, September 2022

[0006] Non-Patent Document 2 assumes a situation in which an object that obstructs the vehicle's travel path is present, and implements control aimed at achieving a quality whereby the object can be detected at least once within a certain period from when the object can be recognized to when the vehicle can stop. Therefore, when the period from when the object can be recognized to when the vehicle can stop is short (for example, when an object suddenly appears), the period becomes very short. The shorter the period, the lower the possibility of improving network quality, so it may not be possible to increase the bit rate to a level where the quality can be recognized.

[0007] The present invention has been made in consideration of the above points, and aims to improve the safety of remote automated driving systems.

[0008] In order to solve the above problem, the information processing device has a recognition rate estimation unit that is configured to, when a vehicle traveling at a certain speed takes a video at a certain point and transmits the video to a monitoring source via a communication network at a certain bit rate, estimate an object recognition rate at the monitoring source from the video based on the bit rate and the speed for a position away from the certain point by a stopping distance at that speed.

[0009] This can improve the safety of remote automated driving systems.

[0010] Fig. 1 is a diagram showing an example of the configuration of a remote automated driving system according to an embodiment of the present invention. Fig. 2 is a diagram showing an example of the hardware configuration of an information processing device 100 according to an embodiment of the present invention. Fig. 3 is a diagram showing an example of the functional configuration of an information processing device 100 according to an embodiment of the present invention. Fig. 4 is a flowchart for explaining an example of a processing procedure executed by the information processing device 100. Fig. 5 is a flowchart for explaining an example of a processing procedure of a deceleration control process. Fig. 6 is a flowchart for explaining an example of a processing procedure of an acceleration control process.

[0011] In this embodiment, a method is disclosed for controlling the driving speed in addition to the bit rate of the monitoring video from the vehicle (hereinafter simply referred to as "video") in order to maintain a state in which a remote monitor / operator can detect dangerous objects and the like from the monitoring video. Note that a remote automated driving system is a system in which a monitor / operator located remotely from a vehicle such as an automobile can perform driving operations.

[0012] It is known that the quality of wireless networks used to distribute video data from vehicles varies depending on the area, as they are strongly affected by obstacles and the surrounding environment. It is also known that the object recognition rate is higher even at lower bit rates when the driving speed is slower.

[0013] Taking these characteristics into consideration, when a driving point does not have sufficient throughput, the system reduces the vehicle's stopping distance by slowing down and delaying the timing when object detection is required, enabling monitoring and operation using video from a different area with sufficient network quality.In addition, by reducing the driving speed, a sufficient recognition rate can be maintained even at a low bit rate, making it possible to deliver video sufficient for object recognition even when network quality is not sufficiently improved.

[0014] Specifically, the bit rate at which video can be transmitted is determined based on an estimate of the network throughput at the driving location and time (or either the driving location or the time), the object recognition rate is estimated from the transmission bit rate and driving speed at each location, and the driving speed and acceleration are determined so that the estimated object recognition rate is equal to or greater than a predetermined threshold.

[0015] In this embodiment, the object recognition rate refers to the probability that a monitor can recognize an object such as a dangerous object that may obstruct driving from a monitored image.

[0016] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0017] 1 is a diagram showing an example of the configuration of a remote automated driving system according to an embodiment of the present invention. In FIG. 1, a vehicle 10 is connected to a monitoring device 20 via a communication network including a wireless access network r1.

[0018] The vehicle 10 is an autonomous vehicle that is traveling autonomously and is equipped with a surveillance camera that captures images of its surroundings and the direction in which it is traveling. The vehicle 10 transmits images captured by the surveillance camera to a monitoring device 20 via a wireless access network r1. The vehicle 10 includes an information processing device 100. The information processing device 100 is a device that executes processing to enable continuous monitoring of the vehicle 10 based on images from the vehicle 10.

[0019] The monitoring device 20 is one or more computers that receive and display images transmitted from the vehicle 10. A remote monitor / operator of the vehicle 10 watches the displayed images to monitor the running state of the vehicle 10 and perform driving operations as necessary.

[0020] In this embodiment, the description focuses on a single autonomous vehicle, that is, vehicle 10, but the processing described below may be executed in parallel for a plurality of autonomous vehicles.

[0021] 2 is a diagram showing an example of the hardware configuration of the information processing device 100 according to the embodiment of the present invention. The information processing device 100 in FIG. 2 includes an auxiliary storage device 102, a memory device 103, a processor 104, a communication device 105, and the like.

[0022] The program that realizes the processing in the information processing device 100 is installed in the auxiliary storage device 102. However, the program does not necessarily have to be installed from the recording medium 101, but may be downloaded from another computer via a network. The auxiliary storage device 102 stores the installed program as well as necessary files, data, etc.

[0023] When an instruction to start a program is received, the memory device 103 reads and stores the program from the auxiliary storage device 102. The processor 104 is a CPU or a GPU (Graphics Processing Unit), or a CPU and a GPU, and executes functions related to the information processing device 100 in accordance with the program stored in the memory device 103. The communication device 105 performs communication via the wireless access network r1.

[0024] 3 is a diagram showing an example of the functional configuration of the information processing device 100 according to the embodiment of the present invention. In FIG. 3, the information processing device 100 includes a throughput estimation unit 11, a vehicle information acquisition unit 12, a speed determination unit 13, a driving control unit 15, and a video transmission unit 16. Each of these units is realized by a process executed by a processor 104 of one or more programs installed in the information processing device 100.

[0025] The throughput estimation unit 11 estimates the throughput of the wireless access network r1 at each point and each time that the vehicle 10 is scheduled to pass through. Note that the vehicle 10 is traveling by autonomous driving, and the future route is assumed to be known. Therefore, each point and each time that the vehicle 10 is scheduled to pass through can also be predicted. Note that the throughput estimation may be performed only for the points or only for the times.

[0026] The vehicle information acquisition unit 12 acquires the traveling speed of the vehicle 10 .

[0027] The speed determination unit 13 receives an estimated throughput value for the vehicle 10 at each location and time, and the vehicle's traveling speed, as input, and determines a traveling speed and acceleration (including deceleration) that enable continuous monitoring, as well as a video bit rate. However, since the actual traveling speed of the vehicle 10 is determined by the vehicle's autonomous driving algorithm, it is assumed that the speed and acceleration information determined by the speed determination unit 13 is input to the driving control unit 15. Note that "enabling continuous monitoring" means that, at the traveling speed of the vehicle 10, an object recognition rate at a position a stopping distance away from the location of the vehicle 10 is equal to or greater than a threshold. To estimate such an object recognition rate, the speed determination unit 13 includes an object recognition rate estimation unit 14. When the vehicle 10 traveling at a certain speed captures video at a certain location and transmits the video to the monitoring source (monitoring device 20) at a certain bit rate via the wireless access network r1, the object recognition rate estimation unit 14 estimates an object recognition rate at the monitoring source based on the video for a position a stopping distance away from the certain location at that speed.

[0028] The driving control unit 15 determines the speed and acceleration of the vehicle 10 based on whether an object can be recognized in the monitoring video and the status of autonomous driving. For example, it is conceivable to adopt the smaller of the speed calculated by the driving control unit 15 and the speed determined by the speed determination unit 13.

[0029] The video transmitting unit 16 changes the video bit rate to the bit rate determined by the rate determining unit 13 and transmits the monitoring video.

[0030] The functions of the information processing device 100 may be provided by the monitoring device 20. That is, the monitoring device 20 may have some or all of the functions of the information processing device 100.

[0031] The following describes the processing procedure executed by the information processing device 100. Fig. 4 is a flowchart for explaining an example of the processing procedure executed by the information processing device 100. The processing procedure in Fig. 4 is repeated every control period (p seconds).

[0032] In step S101, the speed determination unit 13 determines the current position (l ｎｏｗ ) to the point (l ｎｅｘｔ) until the current speed (V ｎｏｗ ) stopping distance (sd ｎｏｗ ) is calculated.

[0033] The control period p seconds is a preset value. ｎｅｘｔ can be calculated from Equation 1.

[0034]

[0035] Here, V ｎｏｗ The vehicle information acquisition unit 12 collects the stopping distance (sd ｎｏｗ ) is calculated using the following formula 2.

[0036]

[0037] Here, RT indicates the reaction time required for an emergency stop after a person recognizes an object, and this value is set in advance. E is the sum of the delay time due to image processing within the vehicle 10, the delay time on the network, and the delay time until the control commanded by the monitor reaches the vehicle 10. This value can be calculated from the delay time of past (most recent) image information. Alternatively, an average delay time may be set in advance to E. f is the friction coefficient, and is a value that changes depending on the road surface conditions, etc. Values ​​corresponding to road surface conditions are stored in advance as a table, and the value of f is set appropriately according to the road surface conditions. Note that the road surface conditions are l ｎｏｗ From ｎｅｘｔ Area up to (l ｎｏｗ and l ｎｅｘｔ , and l ｎｏｗ From ｎｅｘｔ The coefficient of friction f is determined by estimating it from the weather forecast for the area (including the route to the destination). For example, two types of coefficient of friction f may be prepared in advance: one for when the road is wet due to rain and one for when it is not wet. Note that if it is difficult to estimate the road surface conditions, it can also be set manually.

[0038] Next, the speed determination unit 13 ｎｏｗ From ｎｅｘｔ The throughput (Th ｎｏｗ , ..., Th ｎｅｘｔ) is estimated using a throughput estimation technique (S102). At this time, the throughput may be estimated based on information that takes into account the time when the vehicle 10 passes each point in addition to the location (point), or the throughput may be estimated only for the passing time without considering the point. In other words, it is sufficient to estimate the throughput for each point or the time when the vehicle 10 passes each point, or for a set of each point and the passing time of each point. The passing time of each point is ｎｏｗ The distance from the point to the point is V ｎｏｗ The throughput can be obtained by adding the value divided by the current time. The throughput can be estimated for each area and time using a technique such as that in Reference 1. In this embodiment, it is assumed that the throughput changes by a few meters. ｎｏｗ From ｎｅｘｔ The throughput may vary at multiple points along the route.

[0039] Next, the speed determination unit 13 ｎｏｗ From ｎｅｘｔ The lowest throughput value (Th ｎｏｗ・ｍｉｎ ) is substituted into Th in the following equation 3 to calculate the bit rate br, and the br is set to the smallest bit rate (br ｎｏｗ・ｍｉｎ ) (S103).

[0040]

[0041] Here, α is a safety margin that takes into account short-term fluctuations in throughput and estimation errors, and is set to a value of 1 or less in advance.

[0042] Next, the object recognition rate estimation unit 14 calculates the br ｎｏｗ・ｍｉｎ and V ｎｏｗ Based on this, the object recognition rate at the point a stopping distance away (sd) is estimated using a technique such as that in Reference 2 (a model for estimating the object recognition rate from the distance to the object, the bit rate, and the traveling speed) (S104). ｎｏｗ From ｎｅｘｔ The lowest object recognition rate (R ｎｏｗ) Any method for estimating the object recognition rate can be used as long as it uses video quality such as the video bit rate and speed information of the vehicle 10. When using an estimation method that requires input of information other than the bit rate (for example, frame rate or resolution), these values ​​can also be estimated. For example, setting values ​​of the frame rate and resolution for each bit rate can be stored in a table in advance, and these values ​​can be used.

[0043] Next, the speed determination unit 13 determines whether R ｎｏｗ is a preset threshold (R ｔａｒｇｅｔ ) (S105). ｔａｒｇｅｔ It is sufficient that the value is determined appropriately in advance by the operator, taking safety into consideration.

[0044] R ｎｏｗ is R ｔａｒｇｅｔ If it is less than this (Yes in S105), the speed determination unit 13 determines that the current speed cannot be monitored safely until the next control cycle, and executes deceleration control processing (S106).

[0045] On the other hand, R ｎｏｗ is R ｔａｒｇｅｔ If the above condition is met, the speed determination unit 13 determines that monitoring can be safely performed up to the next control cycle even at the current speed, and executes acceleration control processing to determine whether acceleration is possible (S107).

[0046] Next, the details of step S106 will be described. FIG. 5 is a flowchart for explaining an example of the processing procedure of the deceleration control processing. In the deceleration control processing, the speed determination unit 13 determines the current position (l ｎｏｗ ) to the stopping distance sd ｎｏｗ The point just ahead (l ｎｏｗ +sd ｎｏｗ ) is searched for a speed at which the object recognition rate is equal to or greater than a threshold.

[0047] In step S201, the speed determination unit 13 calculates a plurality of candidates for the speed after deceleration. The candidates are called "candidate speeds," and a set of the candidate speeds is called a candidate speed group. For example, the speed determination unit 13 calculates a current speed (V ｎｏｗ ) is subtracted from the predetermined lower limit speed (V ｄｏｗｎｌｉｍ) or less, and V ｄｏｗｎｌｉｍ The larger subtraction result is set as the candidate speed. Here, the lower limit speed is, for example, 0 [m / s]. As a result, a group of candidate speeds at regular speed intervals is obtained. Specifically, V ｎｏｗ = 50 and constant rate = 10, then {40, 30, 20, 10} is obtained as the candidate rate set.

[0048] Next, the speed determination unit 13 repeats the loop process from step S202 onwards for each candidate speed. At this time, each candidate speed is processed in descending order. The candidate speed being processed in the loop process is referred to as "speed V ｄ "

[0049] In step S202, the speed determination unit 13 determines the speed V ｄ Stopping distance (sd ｄ ) is calculated using the following equation 4.

[0050]

[0051] speed V ｄ The stopping distance at ｄ If l ｎｏｗ +sd ｎｏｗ In order to be able to stop at the object at the point, the point shown in the following formula 5 (i.e., l ｎｏｗ +sd ｎｏｗ SD than the point ｄ (the point just before) ｄ You need to slow down to.

[0052]

[0053] Therefore, the speed determination unit 13 determines the speed V at the point of Equation 5. ｄ The deceleration D is calculated based on Equation 6 so that the vehicle decelerates to a speed of 0.05 m / s (S203). It is known that if D is equal to or greater than a certain value, passengers feel uncomfortable. Therefore, the speed determination unit 13 may impose a limit on the maximum value of D. That is, if D exceeds the maximum value, the maximum value may be set as D. When imposing a limit on the maximum value of D, it is necessary to impose a limit on the maximum speed due to the time required for deceleration.

[0054]

[0055] It is necessary to take the control period (p) into consideration in the deceleration control process. ｄ Based on whether the required time for deceleration to reach the destination point l after the control period (p) is equal to or longer than p, ｎｅｘｔ Specifically, as shown in the following equation 7, when the required time for deceleration exceeds p, the speed determination unit 13 calculates l ｎｏｗ From velocity V ｄ The point where the vehicle has traveled the distance until it slows down is called l ｎｅｘｔ On the other hand, if the required time for deceleration is equal to or less than p, the speed determination unit 13 further calculates the difference time (the time obtained by subtracting the required time for deceleration from p) after deceleration as the speed V ｄ The point advanced by l ｎｅｘｔ Let's say.

[0056]

[0057] Next, the speed determination unit 13 ｎｏｗ From ｎｅｘｔ At a plurality of points between V and V, the throughput at each passing time is estimated using a throughput estimation technique such as that in Reference 1 (S205). ｎｏｗ The time required to reach each point based on the speed and deceleration can be obtained by adding the current time to the time.

[0058] Next, the speed determination unit 13 ｎｏｗ From ｎｅｘｔ The throughput estimated for each of the plurality of points between is substituted into Th in Equation 3 to determine (calculate) the transmission bit rate at each point (S205).

[0059] Next, the object recognition rate estimation unit 14 ｎｏｗ From ｎｅｘｔ The transmission bit rate calculated for each point between ｄBased on the estimated speed (taking into account the deceleration to the target point) and the estimated speed, the object recognition rate for a position that is the stopping distance from the target point is estimated for each target point using a technique such as that in Reference 2 (S207). Note that since the speed at each point is different, the stopping distance at each point is also different.

[0060] Next, the speed determination unit 13 sets the minimum value of the object recognition rates estimated for each point as a threshold value (R ｔａｒｇｅｔ ) (S208). ｔａｒｇｅｔ If the current V is less than the predetermined value (Yes in S208), the speed determination unit 13 repeats the loop process. ｄ Therefore, the lower candidate speed is V ｄ Then, step S202 and subsequent steps are executed. ｔａｒｇｅｔ If it is equal to or greater than this (No in S208), the speed determination unit 13 ｄ In this case, the speed determination unit 13 determines (decides) that the vehicle will be decelerated at a deceleration rate D (S209). ｄ The speed determination unit 13 inputs the speed and deceleration D to the driving control unit 15. The speed determination unit 13 also inputs the transmission bit rate corresponding to the minimum value of the object recognition rate to the video transmission unit 16. The driving control unit 15 ｄ The speed and acceleration of the vehicle 10 are controlled with reference to the speed and deceleration D. The video transmission unit 16 transmits the monitoring video at the input transmission bit rate.

[0061] Next, the details of step S107 in Fig. 4 will be described. Fig. 6 is a flowchart for explaining an example of the processing procedure of the acceleration control processing. In the acceleration control processing, the speed determination unit 13 determines the current position (l ｎｏｗ ) to the destination point after acceleration ends (l ｎｅｘｔ ) and if an object at the stopping distance (sd) can be recognized from each point, it is determined that acceleration is possible.

[0062] In step S301, the speed determination unit 13 calculates a plurality of candidates for the speed after acceleration. The candidates are called "candidate speeds," and a set of the candidate speeds is called a candidate speed group. For example, the speed determination unit 13 calculates a current speed (V ｎｏｗ ) and a constant speed is added to the upper limit speed (Vｕｐｌｉｍ ) and repeat the process until V ｕｐｌｉｍ The results of the following additions are used as candidate speeds. Here, the upper speed limit is generally the legal speed limit, but it is necessary to limit it to a specified upper speed limit for the convenience of the deceleration control process. The limit of deceleration is determined from the viewpoint of safety and comfort, and if the speed becomes too high, it may not be possible to decelerate sufficiently. As a result, a group of candidate speeds at regular speed intervals is obtained. Specifically, V ｎｏｗ = 50, constant speed = 10, upper limit speed (V ｕｐｌｉｍ ) is 100, {60, 70, 80, 90, 100} is obtained as a set of candidate speeds. The candidate speeds are sorted in ascending order.

[0063] Next, the speed determination unit 13 repeats the loop process from step S302 onwards for each candidate speed. At this time, each candidate speed is processed in ascending order. The candidate speed being processed in the loop process is referred to as "speed V ａｃ "

[0064] In step S302, the speed determination unit 13 determines whether V ｎｏｗ From velocity V ａｃ The distance d to reach the target point is calculated using the following equation 8.

[0065]

[0066] Here, a indicates acceleration, and is set in advance to a value that is comfortable for the passenger.

[0067] Next, the speed determination unit 13 ｎｏｗ l at a distance d from ｎｅｘｔ At a plurality of points up to , the throughput at each passing time is estimated using a throughput estimation technique such as that in Reference 1 (S303). ｎｏｗ and the time it takes to reach each point based on the acceleration a can be obtained by adding the current time to the time.

[0068] Next, the speed determination unit 13 ｎｏｗ From ｎｅｘｔThe throughput estimated for each of the multiple points between is substituted into Th in Equation 3 to determine (calculate) the transmission bit rate at each point (S304).

[0069] Next, the object recognition rate estimation unit 14 compares the calculated transmission bit rate with ｎｏｗ l at a distance d from ｎｅｘｔ Based on the speed at each point between the point and the target point, the object recognition rate is estimated for each point at a position a stopping distance away from the point (S305) using a technique such as that in Reference 2. In this case, the speed is not the speed at each point taking into account acceleration, but the speed after safely accelerating (V ａｃ In that case, the minimum bit rate and V are calculated from the minimum estimated throughput value in the section. ａｃ The object recognition rate can be calculated from

[0070] Next, the speed determination unit 13 sets the minimum value of the object recognition rates estimated for each point as a threshold value (R ｔａｒｇｅｔ ) (S306). If the minimum value is R ｔａｒｇｅｔ If the speed is equal to or greater than the specified speed (Yes in S306), the speed determination unit 13 sets the specified speed to V ａｃ (S307), and the loop is repeated to see if the speed can be further increased. Therefore, the next candidate speed is V ａｃ Then, step S302 and subsequent steps are executed. The initial value of the command speed is V ｎｏｗ Let's leave it at that.

[0071] On the other hand, the minimum value is R ｔａｒｇｅｔ If it is less than V (No in S306), the speed determination unit 13 ａｃ In this case, the speed determining unit 13 determines that it is dangerous to accelerate to the previous V and determines that the vehicle should travel at the current instructed speed (S308). ａｃ (However, the V to be processed ａｃ If V is the first of the candidate speed group, the speed determination unit 13 inputs the current speed and acceleration a to the driving control unit 15. The speed determination unit 13 also inputs the transmission bit rate corresponding to the minimum value of the object recognition rate to the video transmission unit 16. The driving control unit 15 ａｃand acceleration a, the speed and acceleration of the vehicle 10 are controlled. The video transmitting unit 16 transmits the monitoring video at the input transmission bit rate.

[0072] Regarding steps S303 to S305, when the object recognition rate estimation technique is determined based on the speed and bit rate, V ａｃ The bit rate required for object recognition when ａｃ The bit rate at which the object recognition rate is equal to or greater than the threshold value for the stopping distance at ｎｏｗ From ｎｅｘｔ It may be possible to check that there is no section between the two where the bit rate is lower than the specified bit rate.

[0073] As described above, according to this embodiment, the vehicle 10 can be controlled based on a traveling speed and acceleration (including deceleration) that enable continuous monitoring under the constraints of unstable throughput, and the monitoring video can be transmitted at a video bit rate within the range of the throughput, thereby improving the safety of the remote automated driving system.

[0074] [References] [1] A. Narayanan et al., "Lumos5G: Mapping and Predicting Commercial mmWave 5G Throughput", IMC '20 [2] Koike et al., "Object Recognition Rate Estimation Model for Exterior Vehicle Surveillance Video Taking into Account Quality Variation", IEICE Technical Report, vol. 123, no. 273, CQ2023-39, pp. 8-13, November 2023. The above describes in detail the embodiments of the present invention, but the present invention is not limited to such specific embodiments, and various modifications and variations are possible within the scope of the gist of the present invention described in the claims.

[0075] REFERENCE SIGNS LIST 10 Vehicle 11 Throughput estimation unit 12 Vehicle information acquisition unit 13 Speed ​​determination unit 14 Object recognition rate estimation unit 15 Driving control unit 16 Video transmission unit 20 Monitoring device 100 Information processing device 102 Auxiliary storage device 103 Memory device 104 Processor 105 Communication device

Claims

1. An information processing device comprising: a recognition rate estimation unit configured to, when a vehicle traveling at a certain speed takes a video at a certain point and transmits the video to a monitoring source via a communication network at a certain bit rate, estimate an object recognition rate at the monitoring source from the video based on the bit rate and the speed for a position away from the certain point by a stopping distance at the speed.

2. An information processing device comprising: a speed determination unit configured to determine, for a vehicle that transmits video captured at a certain point while traveling to a monitoring source via a communication network, the traveling speed at which the object recognition rate from the video at the monitoring source becomes equal to or exceeds a threshold value for a position away from the certain point by the stopping distance of the vehicle, based on the throughput estimated for the communication network for the certain point or the time at which the vehicle passes the certain point, or for the combination of the certain point and the time.

3. An information processing device comprising: a speed determination unit configured to determine, for a vehicle that transmits video captured at a certain point while traveling to a monitoring source via a communication network, a traveling speed at which the object recognition rate from the video at the monitoring source becomes equal to or exceeds a threshold value for a position away from the certain point by the stopping distance of the vehicle, based on the throughput of the communication network estimated for the certain point or the time when the vehicle passes the certain point, or for the combination of the certain point and the time; and a driving control unit configured to control the speed of the vehicle based on the traveling speed.

4. An information processing device comprising: a speed determination unit configured to determine, for a vehicle that transmits video captured at a certain point while traveling to a monitoring source via a communication network, a traveling speed at which an object recognition rate from the video at the monitoring source becomes equal to or exceeds a threshold value for a position away from the certain point by the stopping distance of the vehicle, based on an estimated throughput for the communication network for the certain point or the time when the vehicle passes the certain point, or for the combination of the certain point and the time; and a video transmission unit configured to transmit video captured by the vehicle, whose speed is controlled based on the traveling speed, at a bit rate based on the throughput.

5. The information processing device according to claim 1, further comprising a speed determination unit configured to determine a driving speed at which the object recognition rate is equal to or greater than a threshold value based on the stopping distance and the bit rate based on the throughput estimated for the communication network at the certain point.

6. The information processing device according to claim 5, further comprising: a driving control unit configured to control the speed of the vehicle based on the traveling speed.

7. The information processing device according to claim 6, further comprising: a video transmission unit configured to transmit, at the bit rate, video captured by the vehicle whose speed is controlled based on the traveling speed.

Images