platooning control device and platooning control method

Abstract

A platooning control device includes: a learning unit configured to perform reinforcement learning based on image information and a feedback signal, and to control relevant vehicles to follow the trajectory of a vehicle ahead based on the result of the reinforcement learning; and a compensation determination unit configured to receive coordinates of control points regarding the trajectory of the vehicle ahead from the vehicle ahead, and to compare the coordinates of the relevant vehicles with the coordinates of the control points to generate the feedback signal. The platooning control device of this disclosure enables relevant vehicles to stably and effectively follow the trajectory of the vehicle ahead.

Description

[0001] Cross-references to related applications

[0002] This application claims priority and benefit to Korean Patent Application No. 10-2022-0107758, filed with the Korean Intellectual Property Office on August 26, 2022, the disclosure of which is incorporated herein by reference. Technical Field

[0003] This disclosure relates to a platooning control device and a platooning control method, wherein reinforcement learning is performed to follow the trajectory of the front vehicle during platooning. Background Technology

[0004] Generally speaking, platooning refers to driving a group of multiple vehicles on a road while sharing driving information and taking into account the external environment.

[0005] In order to perform stable convoy driving, it is essential to maintain the proper distance between vehicles in the convoy and to control the vehicles behind to follow the trajectory of the vehicles in front.

[0006] Autonomous driving systems can perform reinforcement learning about platooning so that autonomous vehicles can take optimal action during platooning.

[0007] Reinforcement learning is a machine learning method used for learning. It involves taking actions and, through trial and error, finding the optimal outcome in the current state. Each action is rewarded, and the learning process continues to maximize these rewards.

[0008] The above description of the background technology is only for the purpose of helping to understand the background of this disclosure, and those skilled in the art will not consider it to correspond to the known prior art. Summary of the Invention

[0009] Therefore, one aspect of this disclosure is to perform reinforcement learning by using control points and image information about the trajectory of the vehicle in front during platooning, so that the relevant vehicle can stably and effectively follow the trajectory of the vehicle in front.

[0010] The technical topics pursued in this disclosure may not be limited to those described above, and those skilled in the art to which this disclosure pertains can clearly understand other technical topics not mentioned through the following description.

[0011] According to one aspect of this disclosure, a platooning control device may include: a learning device configured to perform reinforcement learning based on image information and feedback signals, and control relevant vehicles to follow the driving trajectory of the vehicle in front based on the results of the reinforcement learning; and a compensation determination unit configured to receive coordinates of control points regarding the driving trajectory of the vehicle in front from the vehicle in front, and compare the coordinates of the relevant vehicles with the coordinates of the control points to generate a feedback signal.

[0012] According to another aspect of this disclosure, the platooning control method may include: controlling relevant vehicles to follow the driving trajectory of a vehicle ahead based on the results of reinforcement learning performed on image information and feedback signals; receiving coordinates of control points regarding the driving trajectory of the vehicle ahead from the vehicle ahead; and generating feedback signals by comparing the coordinates of the relevant vehicles with the coordinates of the control points.

[0013] The advantage of this disclosure is that by using reinforcement learning based on control points and image information about the trajectory of the vehicle in front during convoy driving, the relevant vehicles can stably and effectively follow the trajectory of the vehicle in front.

[0014] The beneficial effects obtained from this disclosure may not be limited to those described above, and other effects not mentioned will be clearly understood by those skilled in the art to which this disclosure pertains through the following description. Attached Figure Description

[0015] The above and other aspects, features and advantages of this disclosure will become more apparent from the following detailed description taken in conjunction with the accompanying drawings, wherein:

[0016] Figure 1 This is a block diagram illustrating an exemplary configuration of a platooning control device according to an embodiment of the present disclosure;

[0017] Figure 2 This is a sequence diagram illustrating the process of exchanging information between a vehicle ahead and related vehicles during platooning, according to an embodiment of the present disclosure;

[0018] Figure 3 The images show front and rear views of vehicles traveling in platoons according to embodiments of the present disclosure;

[0019] Figure 4 An exemplary process for generating control points about the trajectory of a vehicle ahead, according to an embodiment of this disclosure, is shown;

[0020] Figure 5 This is a flowchart illustrating the process of performing reinforcement learning feedback based on control points about the driving trajectory of the vehicle ahead, according to an embodiment of the present disclosure;

[0021] Figure 6The process of performing feedback based on the coordinates of relevant vehicles during platooning, according to an embodiment of this disclosure, is illustrated; and

[0022] Figure 7 This is a flowchart illustrating a process for performing reinforcement learning feedback based on the radio signal strength of a radio signal received from a vehicle ahead, according to an embodiment of this disclosure. Detailed Implementation

[0023] In the following description, the embodiments disclosed herein will be described in detail with reference to the accompanying drawings, and the same or similar elements are given the same and similar reference numerals, so repeated descriptions thereof will be omitted.

[0024] In describing the embodiments disclosed in this specification, detailed descriptions of relevant known technologies may be omitted when they are determined to unnecessarily obscure the essential points of this disclosure. Furthermore, the accompanying drawings are provided merely for ease of understanding of the embodiments disclosed in this specification, and the technical spirit disclosed herein is not limited to the drawings; it should be understood that all variations, equivalents, or substitutions thereof are included within the spirit and scope of this disclosure.

[0025] Terms including ordinal numbers (such as "first", "second", etc.) can be used to describe various elements, but these elements are not limited to these terms. The terms mentioned above are only used to distinguish one element from another.

[0026] Singular expressions can include plural expressions unless they are clearly different in the context.

[0027] As used herein, the expressions “include” or “have” are intended to indicate the presence of the features, quantities, steps, operations, elements, components or combinations thereof mentioned, and should be interpreted as not excluding the possibility of the presence or addition of one or more other features, quantities, steps, operations, elements, components or combinations thereof.

[0028] Figure 1 This is a block diagram illustrating an exemplary configuration of a platooning control device according to an embodiment of the present disclosure.

[0029] like Figure 1 As shown, the platooning control device may include a learning device 100, a compensation determination unit 200, and an inferring neural network device 300.

[0030] According to embodiments of the present disclosure, the platooning control device can use the driving trajectory and image information of the vehicle in front during platooning to perform reinforcement learning, thereby enabling the relevant vehicle to be controlled to stably and effectively follow the driving trajectory of the vehicle in front.

[0031] The components of the platooning control device will now be described.

[0032] The learning device 100 can correspond to an agent as a target for reinforcement learning about platooning.

[0033] The learning device 100 can perform reinforcement learning through a neural network based on image information and feedback signals; and can output steering control signals, braking control signals and acceleration control signals according to the results of reinforcement learning, so that the relevant vehicle is controlled to follow the driving trajectory of the vehicle in front.

[0034] The image information may include forward image information output from the front camera of the relevant vehicle and rear image information output from the rear camera of the vehicle in front. The forward and rear image information may correspond to the state related to platooning and may reflect the characteristics of the real road along which the relevant vehicles are traveling. Therefore, the learning device 100 can perform reinforcement learning using the forward and rear image information corresponding to the current platooning state, so that even in abnormal platooning situations, the relevant vehicle can be controlled to safely follow the trajectory of the vehicle in front.

[0035] Feedback signals can correspond to rewards related to reinforcement learning. More specifically, feedback signals can indicate either positive or negative feedback regarding whether the relevant vehicle follows the trajectory of the vehicle in front. Therefore, the learning device 100 can modify and improve its reinforcement learning strategy based on the feedback signals.

[0036] Steering control signals, braking control signals, and acceleration control signals can correspond to actions related to reinforcement learning and can be generated to perform steering control, braking control, and acceleration control of the relevant vehicle.

[0037] More specifically, the learning device 100 can transmit the control signals required for the driving of the relevant vehicle to the driving-related controller for steering, braking, driving, etc., thereby controlling the driving state of the relevant vehicle.

[0038] For example, the learning device 100 can output a steering control signal to a steering controller (not shown), which is configured to adjust the rotation angle of the steering wheel, thereby controlling the steering angle of the relevant vehicle. It can also output a braking control signal to a brake controller (not shown) configured to adjust the amount of hydraulic braking or to a motor controller (not shown) configured to adjust the amount of regenerative braking, thereby controlling the braking amount of the relevant vehicle. Furthermore, the learning device 100 can output an acceleration control signal to a powertrain controller (not shown), which is configured to adjust the output torque of an electric motor or engine, thereby controlling the acceleration of the relevant vehicle.

[0039] The compensation determination unit 200 can generate a feedback signal corresponding to the reward related to reinforcement learning based on the steering control signal, braking control signal, and acceleration control signal corresponding to the action related to reinforcement learning.

[0040] In addition, the compensation determination unit 200 can receive the coordinates of the control points of the vehicle ahead regarding the vehicle's driving trajectory, and can compare the coordinates of the relevant vehicle with the coordinates of the control points to generate a feedback signal.

[0041] In this embodiment, a control point can be defined as a feature point used to control the shape of a spline curve corresponding to the trajectory of the vehicle ahead.

[0042] A spline curve can correspond to a smooth curve that expresses the trajectory of a vehicle ahead using a spline function. Depending on the embodiment, the spline curve can correspond to either an interpolating spline curve that extends through the control points or an approximating spline curve that does not extend through intermediate control points. Depending on the embodiment, different configurations can be made regarding whether the approximating spline curve extends through the starting control point and the ending control point.

[0043] A method for an operation compensation determination unit 200 to generate a feedback signal will now be described, assuming that the spline curve corresponding to the trajectory of the vehicle ahead corresponds to an approximate spline curve.

[0044] When the coordinates of the relevant vehicle are located outside the driving lane compared to the coordinates of the control point, the compensation determination unit 200 can determine that the relevant vehicle has deviated from the driving trajectory of the vehicle ahead towards the control point, and can output a feedback signal corresponding to the negative feedback. The driving lane refers to the lane in which the relevant vehicle is currently driving.

[0045] Furthermore, when the distance between the coordinates of the relevant vehicle and the coordinates of the control point exceeds the pre-configured danger distance, the compensation determination unit 200 can determine that the relevant vehicle has deviated from the driving trajectory of the vehicle in front in the opposite direction from the control point, and can output a feedback signal corresponding to the negative feedback.

[0046] When negative feedback is input as the result of a coordinate comparison between the relevant vehicle and the control point, the learning device 100 can control the increase of the braking amount of the relevant vehicle through a braking control signal, and can control the steering angle of the relevant vehicle through a steering control signal to follow the driving trajectory of the vehicle in front.

[0047] Conversely, if the coordinates of the relevant vehicle are located inside the driving lane compared to the coordinates of the control point, and if the distance between the coordinates of the relevant vehicle and the coordinates of the control point is within a pre-configured danger distance, the compensation determination unit 200 can determine that the relevant vehicle stably follows the driving trajectory of the vehicle in front. In this case, the compensation determination unit 200 can output a feedback signal corresponding to the positive feedback.

[0048] Therefore, the compensation determination unit 200 according to this embodiment can provide feedback to the learning device 100 on whether the relevant vehicle is following the driving trajectory of the vehicle in front, based on the coordinates of the control points of the driving trajectory of the vehicle in front, thereby reducing the data size and computational load of the driving trajectory of the vehicle in front.

[0049] Furthermore, the compensation determination unit 200 can generate a feedback signal based on whether the radio signal strength (e.g., received signal strength indication, RSSI) of the radio signal received from the vehicle ahead is within a pre-configured range. According to embodiments, the pre-configured range regarding RSSI can be configured differently.

[0050] The RSSI of a radio signal can indicate the inter-vehicle distance between a relevant vehicle and the vehicle ahead. For example, the compensation determination unit 200 can determine that the higher the RSSI, the shorter the inter-vehicle distance between the relevant vehicle and the vehicle ahead.

[0051] If the RSSI of the radio signal is within a pre-configured range, the compensation determination unit 200 can determine that the relevant vehicle stably maintains the vehicle distance from the vehicle in front, and can output a feedback signal corresponding to the positive feedback.

[0052] Conversely, if the RSSI of the radio signal is not included in the pre-configured range, the compensation determination unit 200 can output a feedback signal corresponding to the negative feedback.

[0053] More specifically, if the RSSI of the radio signal is higher than the upper limit of a pre-configured range threshold, the compensation determination unit 200 can determine that the vehicle distance between the relevant vehicle and the vehicle in front is short, and can output a feedback signal corresponding to the negative feedback. The learning device 100 can control the increase of the braking amount of the relevant vehicle through the braking control signal.

[0054] Conversely, if the RSSI of the radio signal is below a pre-configured lower threshold, the compensation determination unit 200 can determine the vehicle distance between the relevant vehicle and the vehicle ahead and can output a feedback signal corresponding to the negative feedback. The learning device 100 can control the acceleration of the relevant vehicle by using an acceleration control signal.

[0055] Therefore, according to this embodiment, the compensation determination unit 200 can provide feedback to the learning device 100 via the RSSI of the radio signal regarding whether the vehicle distance between the relevant vehicle and the vehicle in front remains stable, thereby controlling the learning device 100 to learn the acceleration or braking characteristics related to the distance to the vehicle in front.

[0056] In relation to implementation, the compensation determination unit 200 corresponds to a controller dedicated to feedback on reinforcement learning of the learning device 100, and for this purpose may include a communication device configured to communicate with another controller or sensor, a memory configured to store operating system, logic commands and input / output information, etc., and at least one processor configured to perform determinations, calculations, etc., required for the corresponding functional control.

[0057] After the reinforcement learning of platooning performed by the learning device 100 has stabilized, the inference neural network device 300 can periodically update the parameters of the neural network included in the learning device 100.

[0058] The inference neural network device 300 can receive forward and rear image information and control the relevant vehicle based on updated parameters to track the driving trajectory of the vehicle in front, without feedback from the compensation determination unit 200. The inference neural network device 300 can output steering control signals, braking control signals, and acceleration control signals, as in the case of the learning device 100, so that the relevant vehicle is controlled to follow the driving trajectory of the vehicle in front.

[0059] Therefore, the inference neural network device 300 can perform steering control, braking control and acceleration control of relevant vehicles using only image information after the reinforcement learning of platooning has stabilized without additional reinforcement learning, thereby reducing the computational load of reinforcement learning of the platooning control device.

[0060] Figure 2 This is a sequence diagram illustrating the process of exchanging information between a vehicle ahead and related vehicles during platooning according to an embodiment of the present disclosure.

[0061] exist Figure 2 In the context, the relevant vehicle R has reference value. Figure 1The configuration described herein is assumed to be such that the preceding vehicle F is traveling in platoon with the associated vehicle R and is supported in communicating with the associated vehicle F directly or via infrastructure.

[0062] The vehicle F in front can be downscaled and compressed to generate rear image information (S101), and the related vehicle R can be downscaled and compressed to generate front image information (S103).

[0063] The vehicle in front, F, can send rear image information and radio signals to the relevant vehicle, R, while the relevant vehicle, R, can send forward image information and radio signals to the vehicle in front, F (S105).

[0064] The vehicle ahead, F, can recover the received forward image information and measure the RSSI of the radio signal received from the related vehicle, R (S107). Similarly, the related vehicle, R, can recover the rear image information and measure the RSSI of the radio signal received from the vehicle ahead, F (S109).

[0065] The vehicle F in front can generate a vision-based trajectory (S111) using image information output from the rear camera and image information received from the relevant vehicle R in front, and can generate the coordinates of control points based on the vision-based trajectory (S113).

[0066] The vehicle ahead, F, can transmit the coordinates of the control point to the relevant vehicle, R (S115).

[0067] The relevant vehicle R can provide feedback on reinforcement learning based on the coordinates of the control point and the measurement of the radio signal RSSI (S117), and can perform steering control, braking control and acceleration control of the relevant vehicle R according to the feedback, so as to follow the driving trajectory of the vehicle F in front (S119).

[0068] Figure 3 The images show front and rear views of vehicles traveling in platoons according to embodiments of the present disclosure.

[0069] refer to Figure 3 The first vehicle in front, F <1> Located in front of the relevant vehicle R, the second vehicle in front F <2> Vehicle F located in the first front <1> The front of the vehicle can be captured by the front camera (FV), and the rear of the vehicle can be captured by the rear camera (RV).

[0070] The learning device 100 for the relevant vehicle R can learn from the forward image information of the relevant vehicle R and the first forward vehicle F. <1> The rear image information is used to determine the overlap between the rear image RV of the first forward vehicle F<1> and the forward image FV captured by the relevant vehicle R, and the degree of overlap between the determined rear image RV and the forward image FV can be used as learning data for reinforcement learning.

[0071] For example, the learning device 100 may determine the degree of overlap based on lanes, the shape of markings on the road surface (e.g., road surface signs), feature point extraction, etc., but this is just an example and is not intended to limit the scope in any way.

[0072] Figure 4 An exemplary process for generating control points about the trajectory of a vehicle ahead, according to an embodiment of this disclosure, is shown.

[0073] refer to Figure 4 The vehicle F in front can generate a vision-based trajectory based on rear image information output from the rear camera and front image information received from the vehicle behind. The vehicle F in front can then use the vision-based trajectory to generate the coordinates of control points regarding the driving trajectory of the vehicle in front.

[0074] Figure 5 This is a flowchart illustrating a process of performing feedback on reinforcement learning based on control points about the driving trajectory of a vehicle ahead, according to an embodiment of the present disclosure.

[0075] exist Figure 5 In this context, it is assumed that the learning device 100 is controlling a relevant vehicle to follow the driving trajectory of the vehicle in front, as a result of reinforcement learning performed based on image information and feedback signals.

[0076] The compensation determination unit 200 can receive the coordinates of control points regarding the driving trajectory of the vehicle in front from the vehicle in front (S201). The platooning control device can generate the driving trajectory of the relevant vehicles by using the coordinates of the control points regarding the driving trajectory of the vehicle in front (S203).

[0077] The compensation determination unit 200 can compare the coordinates of the relevant vehicle with the coordinates of the control point (S205, S211) and generate a feedback signal based on the comparison result (S207, S213).

[0078] The compensation determination unit 200 can first determine whether the coordinates of the relevant vehicle are outside the driving lane compared with the coordinates of the control point (S205).

[0079] When the coordinates of the relevant vehicle are located outside the driving lane compared to the coordinates of the control point ("Yes" in S205), the compensation determination unit 200 can output a feedback signal corresponding to the negative feedback. The learning device 100 can increase the braking amount of the relevant vehicle according to the negative feedback control, and can also control the steering angle of the relevant vehicle (S209).

[0080] When the coordinates of the relevant vehicle are located inside the driving lane compared to the coordinates of the control point ("No" in S205), the compensation determination unit 200 can determine whether the distance between the coordinates of the relevant vehicle and the coordinates of the control point exceeds the pre-configured danger distance (S211).

[0081] When the distance between the coordinates of the relevant vehicle and the coordinates of the control point exceeds the preset danger distance ("Yes" in S211), the compensation determination unit 200 can output a feedback signal corresponding to the negative feedback (S207). The learning device 100 can increase the braking amount of the relevant vehicle according to the negative feedback control, and can also control the steering angle of the relevant vehicle (S209).

[0082] When the distance between the coordinates of the relevant vehicle and the coordinates of the control point is within the preset danger distance ("No" in S211), the compensation determination unit 200 can output a feedback signal corresponding to the positive feedback (S213).

[0083] Figure 6 The process of performing feedback based on the coordinates of relevant vehicles during convoy driving is illustrated according to an embodiment of the present disclosure.

[0084] refer to Figure 6 On the left, the first control point to the fourth control point <1:4> of the driving trajectory of the vehicle F ahead is shown.

[0085] Figure 6 The center corresponds to the case where the coordinates of the relevant vehicle R are located outside the driving lane compared to the coordinates of the second control point <2>. Then, the compensation determination unit 200 can output a feedback signal corresponding to the negative feedback.

[0086] Figure 6 The right side corresponds to the case where the coordinates of the relevant vehicle R are located inside the driving lane compared to the coordinates of the second control point <2>, and the distance from the coordinates of the second control point <2> is within the danger distance. Then, the compensation determination unit 200 can output a feedback signal corresponding to the positive feedback.

[0087] Figure 7 This is a flowchart illustrating a process for performing feedback on reinforcement learning based on RSSI of radio signals received from a vehicle ahead, according to an embodiment of this disclosure.

[0088] exist Figure 7 In this context, it is assumed that the learning device 100 is controlling a relevant vehicle to follow the driving trajectory of the vehicle in front, as a result of performing reinforcement learning based on image information and feedback signals.

[0089] The compensation determination unit 200 can receive radio signals from the vehicle in front (S301) and can measure the RSSI of the radio signals (S303).

[0090] The compensation determination unit 200 can determine whether the RSSI of the radio signal is included in the pre-configured range (S305, S311), and can output a feedback signal corresponding to one of the positive feedback and negative feedback based on the determination result (S307, S313, S317).

[0091] The compensation determination unit 200 can first determine whether the RSSI of the radio signal is lower than the upper limit of a pre-configured range (S305).

[0092] When the RSSI is higher than the upper limit of the pre-configured range threshold ("No" in S305), the compensation determination unit 200 can output a feedback signal corresponding to the negative feedback (S307). The learning device 100 can increase the braking amount of the relevant vehicle according to the negative feedback control (S309).

[0093] When RSSI is lower than the upper limit of the pre-configured range threshold ("Yes" in S305), the compensation determination unit 200 can determine whether RSSI is higher than the lower limit of the pre-configured range threshold (S311).

[0094] When the RSSI falls below the lower limit of a pre-configured threshold ("No" in S311), the compensation determination unit 200 can output a feedback signal corresponding to the negative feedback (S313). The learning device 100 can increase the acceleration of the relevant vehicle according to the negative feedback control (S315).

[0095] When RSSI is higher than the lower limit of the pre-configured range threshold ("Yes" in S311), the compensation determination unit 200 can output a feedback signal corresponding to the positive feedback (S317).

[0096] The present disclosure as described above can be implemented as code in a computer-readable medium that records a program. Computer-readable media include all types of recording devices that store data readable by a computer system. Examples of computer-readable media include hard disk drives (HDDs), solid-state drives (SSDs), silicon disk drives (SDDs), read-only memory (ROM), random access memory (RAM), compact disc read-only memory (CD-ROM), magnetic tape, floppy disks, optical data storage devices, etc. Furthermore, the above detailed description should not be construed as restrictive, but should be considered in all respects in an illustrative sense. The scope of this disclosure should be determined by a reasonable interpretation of the technical solutions disclosed herein, and all changes and modifications within the equivalent scope of this disclosure fall within the scope of this disclosure.

Claims

1. A platooning control device, comprising: The learning device is configured to perform reinforcement learning based on image information and feedback signals, and control relevant vehicles to follow the driving trajectory of the vehicle in front based on the result of the reinforcement learning. and The compensation determination unit is configured to receive coordinates of control points relating to the driving trajectory of the vehicle ahead from the vehicle ahead, and compare the coordinates of the relevant vehicle with the coordinates of the control points to generate the feedback signal. The image information includes forward image information output from the front camera of the relevant vehicle and rear image information output from the rear camera of the vehicle in front. The learning device determines the overlap between the rear image of the vehicle in front and the front image of the related vehicle based on the front image information and the rear image information, and determines the degree of overlap between the rear image and the front image as learning data for the reinforcement learning based on the determination result.

2. The platooning control device according to claim 1, wherein the feedback signal indicates one of positive and negative feedback regarding whether the relevant vehicle follows the driving trajectory of the vehicle in front.

3. The platooning control device according to claim 1, wherein the learning device performs one or more of steering control, braking control and acceleration control of the relevant vehicle based on the feedback signal, such that the relevant vehicle follows the driving trajectory of the vehicle in front.

4. The platooning control device according to claim 1, wherein the control point corresponds to a point for controlling the shape of a spline curve corresponding to the driving trajectory of the vehicle in front.

5. The platooning control device according to claim 1, wherein When the coordinates of the relevant vehicle are located outside the driving lane compared to the coordinates of the control point, the compensation determination unit outputs a feedback signal corresponding to the negative feedback.

6. The platooning control device according to claim 5, wherein, When the feedback signal is output, the learning device increases the braking amount of the relevant vehicle and controls the steering angle of the relevant vehicle in order to follow the driving trajectory of the vehicle in front.

7. The platooning control device according to claim 1, wherein, When the distance between the coordinates of the relevant vehicle and the coordinates of the control point exceeds the pre-configured danger distance, the compensation determination unit outputs a feedback signal corresponding to the negative feedback.

8. The platooning control device according to claim 1, wherein, When the coordinates of the relevant vehicle are located inside the driving lane compared to the coordinates of the control point, and when the distance between the coordinates of the relevant vehicle and the coordinates of the control point is within a pre-configured danger distance, the compensation determination unit outputs a feedback signal corresponding to the positive feedback.

9. The platooning control device according to claim 1, wherein the compensation determination unit outputs a feedback signal corresponding to one of positive feedback and negative feedback based on whether the received signal strength of the radio signal received from the vehicle ahead is within a pre-configured range.

10. The platooning control device according to claim 9, wherein, When the received signal strength is within the pre-configured range, the compensation determination unit outputs a feedback signal corresponding to the positive feedback.

11. The platooning control device according to claim 9, wherein, When the received signal strength is higher than the upper limit of the threshold of the pre-configured range, the compensation determination unit outputs a feedback signal corresponding to the negative feedback, and when the feedback signal is output, the learning device increases the braking amount of the relevant vehicle.

12. The platooning control device according to claim 9, wherein, When the received signal strength is lower than the lower limit of the pre-configured range threshold, the compensation determination unit outputs a feedback signal corresponding to the negative feedback; and when the feedback signal is output, the learning device increases the acceleration of the relevant vehicle.

13. The platooning control device of claim 1 further includes an inferential neural network device configured to update parameters of the neural network included in the learning device to receive the image information based on the updated parameters and control the associated vehicles to follow the driving trajectory of the vehicle in front.

14. A platooning control method, comprising: The relevant vehicle is controlled based on the results of reinforcement learning performed on image information and feedback signals to follow the driving trajectory of the vehicle in front. Receive the coordinates of control points regarding the driving trajectory of the vehicle ahead; and The feedback signal is generated by comparing the coordinates of the relevant vehicle with the coordinates of the control point. The image information includes forward image information output from the front camera of the relevant vehicle and rear image information output from the rear camera of the vehicle in front. The overlapping portion between the rear image of the vehicle in front and the front image of the related vehicle is determined based on the front image information and the rear image information, and the degree of overlap between the rear image and the front image is determined as learning data for the reinforcement learning based on the determined result.

15. The platooning driving control method according to claim 14, wherein the control point corresponds to a point for controlling the shape of a spline curve corresponding to the driving trajectory of the vehicle in front.

16. The platooning control method according to claim 14, further comprising: Receive radio signals from the vehicle ahead; Determine whether the received signal strength of the radio signal is within a pre-configured range; and Based on the determined result, a feedback signal corresponding to one of the positive and negative feedback is output.

17. A non-transitory computer-readable recording medium having a program recorded thereon for performing the platooning control method of claim 14.

Images