A method and system for switching the human-machine control driving right of an inland water surface ship

Abstract

The present application discloses a method and system for switching the human-machine control driving right of inland water surface ships. The method includes: obtaining comprehensive ship navigation status information; determining whether there is a navigation risk for the ship according to the comprehensive ship navigation status information; when there is a navigation risk for the ship, calculating the risk value of the ship; and switching the driving right according to the risk value through a preset human-machine co-driving strategy based on the Nash game theory. The present invention determines the navigation risk of the ship by obtaining the comprehensive ship navigation status information. When there is a navigation risk for the ship, it calculates the risk value of the ship; according to the risk value, through a preset human-machine co-driving strategy based on the Nash game theory, and uses the confidence matrix to update in real time to achieve a smooth transition of the driving right from the driver to the intelligent control system, so as to improve the ship safety while ensuring that the driver is always in the loop, and ensure the safety of personnel and ships during the driving right switching process of human-machine dual driving of the ship.

Description

Technical Field

[0001] The present invention relates to the technical field of ship navigation control, and particularly to a method and system for switching the human-machine control driving rights of inland water surface ships. Background Art

[0002] During the process of inland waterway shipping, driver operation errors are the main cause of traffic accidents. With the continuous development of intelligent driving technology, the intelligent driving system can take over the driving system when a dangerous situation occurs or the driver's behavior is inaccurate, correct the driver's incorrect driving instructions or directly take over the driving system, avoid the occurrence of traffic accidents, and thus ensure the safety of navigation.

[0003] In the existing inland water surface ship during the navigation control process, in order to avoid traffic accidents caused by driver operation errors, when the navigation risk is relatively small, the driver has full control of the driving authority; when the navigation risk is relatively large, the intelligent navigation system has full control of the driving authority. However, the existing driving system lacks the conversion operation of human-machine co-driving during the transformation between the driver and the intelligent system, pays less attention to the smoothness and flexibility during the driving right switching process between the driver / intelligent driving system, and cannot update the weight range of the human-machine driving rights in real time during the driving right switching, resulting in poor smoothness of the transition of the ship's driving right from the driver to the intelligent driving system, and cannot ensure that the driver is always in the loop during the switching, affecting the safety during the ship's driving right switching.

[0004] Therefore, a method and system for switching the human-machine co-driving control mode of inland water surface ships are needed, which can effectively ensure the driving right switching timing and the smoothness and flexibility during the driving right switching process. Summary of the Invention

[0005] In view of this, it is necessary to provide a method and system for switching the human-machine control driving rights of inland water surface ships to solve the problems of poor smoothness and flexibility when the driving right of the existing inland water surface ship is controlled by the driver to the intelligent driving system, and there are potential safety hazards during the driving right switching process.

[0006] To solve the above problems, the present invention provides a method for switching the human-machine control driving rights of inland water surface ships, including:

[0007] Obtain the comprehensive state information of ship navigation;

[0008] Determine whether there is a navigation risk for the ship according to the comprehensive state information of ship navigation;

[0009] When there is a navigation risk for the ship, calculate the risk value of the ship;

[0010] According to the risk value, switch the driving right through a preset human-machine co-driving strategy based on the Nash game theory.

[0011] Furthermore, the comprehensive ship navigation status information includes: the ship's own status information, the navigation scenario information, and the driver's control command information.

[0012] Furthermore, determining whether there is a navigation risk for the ship according to the comprehensive ship navigation status information includes:

[0013] Determining the feasible domain of the ship according to the ship's own status information and the navigation scenario information;

[0014] Determining the predicted navigation trajectory of the ship according to the ship's own status information, the navigation scenario information, and the driver's control command information;

[0015] Judging whether the feasible domain of the ship and the predicted navigation trajectory coincide. If there is no coincidence, it is determined that the ship has no navigation risk; if there is a coincidence, it is determined that the ship has a navigation risk.

[0016] Furthermore, the risk value is determined according to the ship's advancing risk situation, the distance risk situation, and the navigation state risk situation.

[0017] Furthermore, according to the risk value, switching the driving right through a preset human-machine co-driving strategy based on the Nash game theory includes:

[0018] When the risk value is within the first preset threshold range, switching the driving right to the driver's full control mode;

[0019] When the risk value is within the second preset threshold range, switching the driving right to the human-machine co-driving control mode;

[0020] When the risk value is within the third preset range, switching the driving right to the full control mode of the intelligent navigation system.

[0021] Furthermore, the preset human-machine co-driving strategy based on the Nash game theory includes:

[0022] Establishing a comprehensive cost function with the driving right allocation coefficients of the driver and the intelligent driving system as independent variables and the sum of the ship operation amount and the error between the current navigation trajectory and the planned trajectory as the dependent variable;

[0023] Taking the minimization of the comprehensive cost function as the goal to determine the optimal ship control sequence;

[0024] Solving the driving right allocation coefficients of the driver and the intelligent driving system corresponding to the optimal ship control sequence based on the Nash game theory.

[0025] Furthermore, smoothing the control output parameters before switching the driving right, and the time of the control output parameters is greater than the preset minimum output time interval.

[0026] The present invention also provides a device for switching the human-machine control driving right of an inland water surface ship, including:

[0027] An integrated status information acquisition module, configured to acquire the integrated status information of the ship's navigation;

[0028] A navigation risk judgment module, configured to determine whether there is a navigation risk for the ship according to the integrated status information of the ship's navigation;

[0029] A risk hazard value calculation module, configured to calculate the risk hazard value of the ship when there is a navigation risk for the ship;

[0030] A driving right switching module, configured to switch the driving right according to the risk hazard value through a preset human-machine co-driving strategy based on the Nash game theory.

[0031] The present invention also provides an intelligent control system for human-machine co-driving on the inland water surface, including a device for switching the human-machine control driving right of an inland water surface ship as described in the above technical solution.

[0032] The present invention also provides an inland water surface ship, including an intelligent control system for human-machine co-driving on the inland water surface as described in the above technical solution.

[0033] Compared with the prior art, the beneficial effects of the present invention include: First, acquiring the integrated status information of the ship's navigation; Second, determining the navigation risk of the ship according to the integrated status information of the ship's navigation; When there is a navigation risk for the ship, calculating the risk hazard value of the ship; Finally, switching the driving right according to the risk hazard value through a preset human-machine co-driving strategy based on the Nash game theory. The method of the present invention determines whether there is a navigation risk for the ship through the acquired integrated status information of the ship's navigation. When there is a navigation risk for the ship, calculating the risk hazard value of the ship; According to the risk hazard value, through a preset human-machine co-driving strategy based on the Nash game theory, using the confidence matrix to update in real time to achieve a smooth transition of the driving right from the driver to the intelligent control system, so as to improve the safety of the ship while ensuring that the driver is always in the loop, and ensure the safety of personnel and the ship during the driving right switching process of human-machine dual driving of the ship. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic flowchart of an embodiment of a method for switching the human-machine control driving right of an inland water surface ship provided by the present invention;

[0035] Figure 2 It is a schematic diagram of the motion force condition of an embodiment of a ship model provided by the present invention;

[0036] Figure 3Overall process framework diagram of an embodiment of the driving right switching provided by the present invention;

[0037] Figure 4 Schematic flow diagram of an embodiment of the human-machine co-driving transition provided by the present invention;

[0038] Figure 5 Overall process block diagram of an embodiment of the human-machine co-driving strategy based on the Nash game theory provided by the present invention;

[0039] Figure 6 Schematic structural diagram of an embodiment of a device for switching the human-machine control driving right of an inland water surface ship provided by the present invention. Detailed implementation manners

[0040] The following will specifically describe the preferred embodiments of the present invention in conjunction with the accompanying drawings. The accompanying drawings form a part of this application and are used together with the embodiments of the present invention to explain the principles of the present invention, rather than to limit the scope of the present invention.

[0041] The embodiment of the present invention provides a method for switching the human-machine control driving right of an inland water surface ship, as Figure 1 shown, Figure 1 is the schematic flow diagram of the method for switching the human-machine control driving right of the inland water surface ship, including:

[0042] Step S101: Obtain the comprehensive state information of the ship's navigation;

[0043] Step S102: Determine the navigation risk of the ship according to the comprehensive state information of the ship's navigation;

[0044] Step S103: When there is a navigation risk for the ship, calculate the risk hazard value of the ship;

[0045] Step S104: According to the risk hazard value, switch the driving right through a preset human-machine co-driving strategy based on the Nash game theory.

[0046] The method for switching the human-machine control driving right of inland water surface ships provided by this embodiment first obtains the comprehensive ship navigation status information; secondly, determines the navigation risk of the ship according to the comprehensive ship navigation status information; when the ship has a navigation risk, calculates the risk value of the ship; finally, according to the risk value, switches the driving right through a preset human-machine co-driving strategy based on the Nash game theory. The method of this embodiment determines whether the ship has a navigation risk through the obtained comprehensive ship navigation status information. When the ship has a navigation risk, calculates the risk value of the ship; according to the risk value, through a preset human-machine co-driving strategy based on the Nash game theory, uses the confidence matrix to update in real time to achieve a smooth transition of the driving right from the driver to the intelligent control system, so as to improve the ship safety while ensuring that the driver is always in the loop, and ensure the safety of personnel and ships during the driving right switching process of human-machine dual driving of the ship.

[0047] As a preferred embodiment, in step S101, the comprehensive ship navigation status information includes: ship own status information, navigation scenario information, and driver control instruction information.

[0048] As a specific embodiment, various sensors of the ship are used to collect the ship own status information, navigation scenario information, and driver control instruction information during navigation.

[0049] The ship own status data includes: ship own maneuverability, actual main engine speed, speed, course, longitude and latitude information. Among them, extracting the ship own maneuverability information includes turning performance, stopping performance, course-keeping performance, etc., which is because considering that the driving difficulty of ships is different under different tonnages, different scales, and different propulsion methods, and the corresponding behaviors of ship drivers are also different; extracting the actual main engine speed of the ship itself is because considering that there may be a certain deviation between the main engine speed of the ship and the actual speed, and the ship driver can correct the current path planning based on experience; extracting speed, course, longitude and latitude information is because when predicting the ship state trajectory and switching the driving right, it is necessary to analyze in combination with the current navigation speed, direction and specific position of the ship.

[0050] The navigation scenario information includes the navigation status information of other ships in the ship navigation waters, as well as the obstacle and water area natural environment information in the water area. Specifically includes: the speed, course, longitude and latitude information of surrounding ships, video target category and distance information, distance and speed information of radar targets, information such as the current water flow velocity and direction, wind speed and direction, visibility in the current sea area or river course, and electronic nautical chart (channel chart) information.

[0051] When determining whether to switch the driving right, considering that the information of other ships in the ship navigation waters has an important impact on the path planning of ship drivers, the speed, course, longitude and latitude information of surrounding ships is obtained through the Automatic Identification System (AIS) of ships; at the same time, the category and distance information of the video image target also need to be obtained through video images. Here, considering that other video obstacles in this water area have an important impact on the path planning judgment of ship drivers, and here, in place of the lookout sailor, 24-hour all-weather monitoring can be achieved. The combination of an infrared camera and an ordinary camera is adopted, which also has a certain reconnaissance ability at night; the distance and speed information of surrounding targets are obtained through radar, and the water flow velocity and direction, wind speed and direction, and visibility information of the current sea area or river are collected through natural environment collection sensors; the information of surrounding waterway shorelines, water depths, bridges, navigation marks, buoys, etc. is obtained through electronic charts.

[0052] The driver control instruction information described includes the instruction information for the driver to control the rudder and the instruction information for adjusting the engine telegraph.

[0053] By comprehensively considering the ship's own state information, navigation scenario information, driver control instruction information, etc., reliable data support can be provided for the subsequent switching process.

[0054] In some embodiments, the ship classifies and stores the collected comprehensive ship navigation state information to generate a ship model database, a navigation scenario database, and a corresponding driver operation database, which can provide a data basis for subsequent research and analysis in this field.

[0055] As a preferred embodiment, in step S102, determining whether the ship has a navigation risk according to the comprehensive ship navigation state information includes:

[0056] Determining the feasible domain of the ship according to the ship's own state information and navigation scenario information;

[0057] Determining the predicted navigation trajectory of the ship according to the ship's own state information, navigation scenario information, and driver control instruction information;

[0058] Judging whether the feasible domain and the predicted navigation trajectory of the ship coincide. If there is no coincidence, it is determined that the ship has no navigation risk; if there is a coincidence, it is determined that the ship has a navigation risk.

[0059] As a specific embodiment, the feasible domain is specifically: calculating the relative distance and speed information with surrounding targets according to the current ship maneuverability, speed, and course. If the relative distance and speed information with surrounding targets are within the preset safety distance threshold and safety speed range, it is determined that the navigation range between the ship and surrounding targets is the feasible domain; the surrounding targets include other ships, islands, river embankments, and reefs on the water area.

[0060] As a specific embodiment, the predicted navigation trajectory is obtained by inputting the ship's own state information, navigation scenario information, and driver's control instruction information into a trained BP neural network. The predicted navigation trajectory is compared with the feasible region of the ship. If the predicted navigation trajectory is within the feasible region, the ship is in a safe navigation state and can continue to navigate in the mode completely controlled by the driver without switching the driving authority; when there is a navigation risk, the driving authority needs to be switched.

[0061] The following uses a specific implementation model and formula to elaborate in detail on the determination of the above-mentioned feasible region, predicted navigation trajectory, and navigation risk.

[0062] As a specific embodiment, the ship model is expressed as:

[0063]

[0064] As Figure 2 shown, Figure 2 is a schematic diagram of the ship model.

[0065] In Equation (1), n is the position and orientation of the ship in the inertial coordinate system; n = [x s y s ψ] T ; x s is the position of the ship in the X-axis direction; y s is the position of the ship in the Y-axis direction; ψ is the heading angle of the ship; v is the speed and angular velocity of the ship in the attached body coordinate system; v = [u s v y r] T ; u s is the speed of the ship in the x-axis direction; v y is the speed of the ship in the y-axis direction; r is the yaw angular velocity of the ship's axis; T is the rotation transformation matrix; M is the inertia matrix; C is the Coriolis centripetal matrix; D is the damping matrix; u is the total input control force and torque of the driver and the intelligent navigation system; U = ζ1 + ζ2 + ζ d ; ζ1 is the input control force and torque of the driver; ζ2 is the input control force and torque of the intelligent navigation system; ζ d is the wind and wave interference force and torque; ζ d = [f du f dv t dr T ; f du is the interference in the x-axis direction generated by the wind and waves; f dv is the interference in the y-axis direction generated by the wind and waves; t dr is the wind and wave interference torque.

[0066]

[0067]

[0068]

[0069]

[0070] c 23 = m 11 U; ψ is the yaw angle of the ship's shaft; f u is the forward control force generated by the ship's actuator; f v is the side-slip control force; t r is the yaw control moment; m ij are the diagonal elements of the ship's inertia matrix, including added hydrodynamic mass, i = 1, 2, 3; j = 1, 2, 3; d ij are the hydrodynamic viscosities in the ship's sway and yaw directions, i = 1, 2, 3; j = 1, 2, 3.

[0071] Since most intelligent driving systems are controlled in the MPC mode, and when using MPC control, it is necessary to determine the state variables of the ship and discretize the state variables and operation variables. Therefore, the feasible region of the ship model is expressed in the form of a state-space equation, specifically as follows:

[0072]

[0073]

[0074] Among them, p1; p2 are coefficient matrices, which can be expressed as:

[0075] p1 = [0 0 1 0 0 0]

[0076] The ship state vector can be expressed as:

[0077] x = [n T v T T

[0078] The output part coefficient matrix is:

[0079]

[0080] After discretization, we can get:

[0081]

[0082] Taking N p as the prediction horizon, let N c be the control horizon with N c ≤ N​p , the predictive equation can be obtained through continuous iteration as follows:

[0083] Y p (k + 1|k) = Fx(k) + L1ΔU1(k) + L2ΔU2(k) (5)

[0084] In Equation (5), ΔU1(k) is the driver input, ΔU2(k) is the intelligent navigation system input, and F, L1, and L2 are control coefficients.

[0085]

[0086]

[0087] Collect the driver's operation instructions through sensors, determine the predicted navigation trajectory of the ship based on the ship's current own state information, navigation scenario information, and driver's operation instruction information, and compare it with the feasible region of the ship calculated above to determine whether there is a navigation risk for the ship.

[0088] As a preferred embodiment, the risk value is determined according to the ship's travel risk situation, distance risk situation, and navigation state risk situation.

[0089] As a preferred embodiment, in step S103, according to the risk value, the driving right is switched through a preset human-machine co-driving strategy based on the Nash game theory, including:

[0090] When the risk value is within the first preset threshold range, the driving right is switched to the driver full control mode;

[0091] When the risk value is within the second preset threshold range, the driving right is switched to the human-machine co-driving control mode;

[0092] When the risk value is within the third preset range, the driving right is switched to the intelligent navigation system full control mode.

[0093] As a specific embodiment, the calculation of the risk value is based on the field theory. Specifically:

[0094] Define the ship's travel risk situation D as:

[0095]

[0096] where z is the type of the ship, t is the displacement of the ship, h is the draft of the ship, v is the ship's speed, and v max is the maximum speed of the ship, and k1 is the weight coefficient.

[0097] Define the distance risk distribution situation D of the ship dDefined as:

[0098]

[0099] Wherein, k2 is a weight coefficient, and d is the distance between the ship and the obstacle.

[0100] The risk distribution situation D of the ship's navigation state ns Is defined as:

[0101] D ns = k3·(e [v·cos(θ) +e [v·sin(θ)] ) (8)

[0102] Wherein, k3 is a weight coefficient, v is the ship's navigation speed, and θ is the included angle between the ship's navigation direction and the obstacle.

[0103] Therefore, the risk value composed of the above three influencing factors, namely the ship's travel risk situation, distance risk situation, and navigation state risk situation, can be expressed as:

[0104] D c = D·D d ·D ns (9)

[0105] Wherein, D c ∈[0, 1].

[0106] Define D c ∈[0, 0.4] as the safe interval. At this time, the ship is in a safe state, and the driver fully masters the driving authority;

[0107] Define D c ∈[0.4, 0.8] as the risk interval. At this time, it enters the risk state, and the driving authority is jointly mastered by the driver and the intelligent navigation system, that is, it enters the human-machine co-driving mode;

[0108] Define D c ∈[0.8, 1.0] as the dangerous state. At this time, the driving authority is fully mastered by the intelligent navigation system.

[0109] To more intuitively illustrate the above switching process, as Figure 3 shown, Figure 3 shows the overall process framework diagram of the driving right switching.

[0110] As a preferred embodiment, the preset human-machine co-driving strategy based on the Nash game theory includes:

[0111] Establish a comprehensive cost function with the driving right allocation coefficients of the driver and the intelligent driving system as independent variables and the sum of the ship's operation amount and the error between the current navigation trajectory and the planned trajectory as the dependent variable;

[0112] Determine the optimal ship control sequence with the goal of minimizing the comprehensive cost function;

[0113] Based on the Nash game theory, solve the driving right allocation coefficients of the driver and the intelligent driving system corresponding to the optimal ship control sequence.

[0114] As a specific embodiment, in the human-machine co-driving mode, through a human-machine co-driving strategy based on the Nash game theory, the smooth transition of the driving right from the driver to the intelligent navigation system is realized by real-time updating of the confidence matrix. Specifically:

[0115] Define the comprehensive cost function of the driver and the intelligent navigation system as:

[0116] V(k)=V1(k)+V2(k); (10)

[0117] Wherein,

[0118]

[0119] The comprehensive cost function is the sum of the error between the current trajectory of the driver / intelligent driving system and the ship's planned trajectory plus the current operation amount of the driver / intelligent driving system. When the value of the comprehensive cost function is smaller, the safety of ship navigation is higher. Therefore, it is necessary to solve the minimum value of the comprehensive cost function, and at this time, the safety of ship navigation is the highest.

[0120] In formula (10),

[0121] T1(k) and T2(k) are the local target trajectories of the driver and the intelligent navigation system, and k represents the update iteration number. Among them, the local target trajectory is the target trajectory considering only one type of control method. Rolling update needs to be performed before each optimization, specifically:

[0122] T1(k + 1)=BT1(k)+Ht1(k + 1)

[0123] T2(k + 1)=BT2(k)+Ht2(k + 1) (11)

[0124] In the formula,

[0125]

[0126] I m is an m-dimensional identity matrix, m is the number of state variables;

[0127]

[0128] Both Q1(k) and Q2(k) are time-varying matrices. The transfer of the driving right between the driver and the intelligent navigation system of the intelligent ship can be realized through the changes of the two confidence matrices q1(k) and q2(k) in Q1(k) and Q2(k).

[0129]

[0130]

[0131] q1(k) is the driver confidence matrix, and q2(k) is the intelligent navigation system confidence matrix.

[0132]

[0133]

[0134] κ1(k) and κ2(k) are parameters related to the driving right allocation, and λ1(k) and λ2(k) are parameters related to the dynamic characteristics, namely the dynamic adjustment coefficients.

[0135] When the ship enters a dangerous state, κ1(k) will gradually decrease and κ2(k) will gradually increase. During this process, the driving authority is jointly controlled by the driver and the intelligent navigation system. The ship will sail according to the navigation route planned by the intelligent navigation system until the dangerous state ends; when the ship gets out of the risk state, κ1(k) will gradually increase and κ2(k) will gradually decrease, and at this time the driving right is gradually returned to the driver.

[0136] In the state of human-machine co-driving, the optimal control sequences of the driver and the control system are related to the control instructions of the other party, and the optimal control sequence can be solved through Nash game.

[0137] Define the error variables of the driver and the intelligent navigation system as ε1(k) and ε2(k):

[0138] ε1(k) = T1(k) - Fx(k) - L2ΔU2(k)

[0139] ε2(k) = T2(k) - Fx(k) - L1ΔU1(k) (12)

[0140] Then there is

[0141]

[0142] The partial derivative of V1(k) with respect to ΔU1(k) is:

[0143]

[0144] V i(k) The solution where the partial derivative with respect to ΔU1(k) is equal to 0 is the optimal control sequence that minimizes the cost function, i.e.:

[0145]

[0146]

[0147] where

[0148]

[0149] represents the optimal control sequence of the i-th (i = 1, 2) control input ΔU1(k), Table ΔU i (k) The set of all possible optimal control sequences. The solution process of the above optimal control sequence set is as Figure 4 shown.

[0150] Based on game theory, the game system composed of the ship steering system described by Equation (2) and the comprehensive cost function described by Equation (12) can be solved. The game system has a unique Nash equilibrium solution if and only if the preset matrix I - S(k) is invertible. The Nash equilibrium solution is:

[0151]

[0152] where, E(k) = [I - S(k)] -1 M(k),

[0153]

[0154] The feedback gains k1(k) and k2(k) of the driver and the intelligent navigation system are:

[0155]

[0156] where, I i is the identity matrix, and i is the number of control inputs for each participant.

[0157] At this time, the Nash equilibrium solution of the optimal control sequence is obtained, that is: the parameters related to the optimal driving right allocation of the driver and the intelligent driving system are obtained. According to the parameters related to the optimal driving right allocation, the safety of the ship's navigation during the human - machine co - driving process can be maximized.

[0158] It can be understood that during the entire switching process, due to the continuous change of input data, k1(k) and k2(k) may oscillate. To ensure the smoothness and stability of the driving right switching, as a preferred embodiment, smoothing processing should be performed before outputting the actual control parameters, and the output time of the control parameters is greater than the preset minimum output time interval.

[0159] To illustrate the above overall process more intuitively, as Figure 5 shown, Figure 5 it shows the overall process block diagram of the human-machine co-driving strategy based on the Nash game theory.

[0160] As a specific embodiment, in order to make the control parameter output smooth, a moving average filter is included in the system:

[0161]

[0162] where t1 represents the window length of the filter, and i = 1, 2.

[0163] To adjust the control parameter output interval, a zero-order hold is included in the system:

[0164]

[0165] t2 is the sampling interval of the zero-order hold.

[0166] Therefore, in the human-machine co-driving mode, the control inputs of the driver and the intelligent navigation system are respectively:

[0167]

[0168]

[0169] As a specific embodiment, in the human-machine co-driving mode, tactile shared control is adopted. When the control system expects to take over the driving right, an additional torque will be applied to the steering gear to control the ship, and the driving intention of the driver will be sensed according to the torque applied by the driver on the steering gear, and then the driver will be assisted through the steering gear torque.

[0170] In the autonomous driving mode, the intelligent navigation system uses the A* algorithm for path planning according to the received environmental information, and uses MPC control to make the ship return to the safe route along the planned trajectory safely and quickly.

[0171] The embodiment of the present invention also provides a device for switching the human-machine control driving right of an inland water surface ship, as Figure 6 shown, Figure 6 it is a schematic structural diagram of a device 600 for switching the human-machine control driving right of an inland water surface ship provided in this embodiment, including:

[0172] An integrated status information acquisition module 601, configured to acquire the comprehensive status information of ship navigation;

[0173] A navigation risk judgment module 602, configured to determine whether there is a navigation risk for the ship according to the comprehensive status information of ship navigation;

[0174] A risk and hazard value calculation module 603, configured to calculate the risk and hazard value of the ship when the ship has a navigation risk;

[0175] A driving right switching module 604, configured to switch the driving right according to the risk and hazard value through a preset human-machine co-driving strategy based on the Nash game theory.

[0176] This embodiment also provides an intelligent human-machine co-driving control system for inland river water surfaces, including the human-machine control driving right switching device for an inland river water surface ship described in the above technical solution.

[0177] This embodiment also provides an inland river water surface ship, including an intelligent human-machine co-driving control system for inland river water surfaces described in the above technical solution.

[0178] A method, device, intelligent human-machine co-driving control system for inland river water surfaces, and inland river water surface ship provided by the present invention. First, obtain the comprehensive ship navigation status information; secondly, determine the navigation risk of the ship according to the comprehensive ship navigation status information; when the navigation risk is greater than a preset risk threshold, switch the driving right through a preset human-machine co-driving strategy based on the Nash game theory.

[0179] The method of the present invention determines the navigation risk of the ship through the obtained comprehensive ship navigation status information, determines whether the ship encounters a dangerous situation through the navigation risk. When the navigation risk exceeds the preset risk threshold, the driving right is switched through a preset human-machine co-driving strategy based on the Nash game theory, so that the driving right is smoothly and gently transitioned from the driver to the intelligent navigation system, and at the same time, the safety of personnel and the ship is ensured during the driving right switching process of the human-machine dual-driving of the ship.

[0180] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for switching the human-machine control driving right of an inland water surface ship, characterized in that, including; obtaining comprehensive ship navigation status information; determining whether there is a navigation risk for the ship according to the comprehensive ship navigation status information; when there is a navigation risk for the ship, calculating the risk hazard value of the ship; switching the driving right according to the risk hazard value through a preset human-machine co-driving strategy based on the Nash game theory; switching the driving right according to the risk hazard value through a preset human-machine co-driving strategy based on the Nash game theory, including: when the risk hazard value is within the first preset threshold range, switching the driving right to the driver full control mode; when the risk hazard value is within the second preset threshold range, switching the driving right to the human-machine co-driving control mode; when the risk hazard value is within the third preset range, switching the driving right to the intelligent navigation system full control mode; the preset human-machine co-driving strategy based on the Nash game theory includes: establishing a comprehensive cost function with the driving right allocation coefficients of the driver and the intelligent driving system as independent variables and the sum of the ship operation amount and the error between the current navigation trajectory and the planned trajectory as the dependent variable; taking minimizing the comprehensive cost function as the goal to determine the optimal ship control sequence; solving the driving right allocation coefficients of the driver and the intelligent driving system corresponding to the optimal ship control sequence based on the Nash game theory; the comprehensive cost function is: In the formula, is the number of update iterations, is the comprehensive cost function, is the error variable of the driver, is the error variable of the intelligent navigation system, and are the parameters related to the driving right allocation, and are the dynamic adjustment coefficients.

2. The method for switching the human-machine control driving right of an inland water surface ship according to claim 1, wherein, the comprehensive ship navigation status information includes: ship self-status information, navigation scenario information and driver control instruction information.

3. A method for switching the human-machine control driving right of an inland water surface ship according to claim 2, characterized in that determining whether there is a navigation risk for the ship according to the comprehensive ship navigation status information, including: determining the feasible domain of the ship according to the ship self-status information and the navigation scenario information; determining the predicted navigation trajectory of the ship according to the ship self-status information, the navigation scenario information and the driver control instruction information; judging whether the feasible domain of the ship and the predicted navigation trajectory coincide. If there is no coincidence, it is determined that the ship has no navigation risk; if there is a coincidence, it is determined that the ship has a navigation risk.

4. A method for switching the human-machine control driving right of an inland water surface ship according to claim 1, characterized in that, the risk hazard value is determined according to the ship's progress risk situation, distance risk situation and navigation status risk situation.

5. A method for switching the human-machine control driving right of an inland water surface ship according to claim 1, characterized in that, smoothing the control output parameters before switching the driving right, and the time of the control output parameters is greater than the preset minimum output time interval.

6. A human-machine control driving right switching device for inland water surface ships, characterized in that, including: a comprehensive status information acquisition module for obtaining comprehensive ship navigation status information; a navigation risk judgment module for determining whether there is a navigation risk for the ship according to the comprehensive ship navigation status information; a risk hazard value calculation module for calculating the risk hazard value of the ship when there is a navigation risk for the ship; a driving right switching module for switching the driving right according to the risk hazard value through a preset human-machine co-driving strategy based on the Nash game theory; switching the driving right according to the risk hazard value through a preset human-machine co-driving strategy based on the Nash game theory, including: when the risk hazard value is within the first preset threshold range, switching the driving right to the driver full control mode; when the risk hazard value is within the second preset threshold range, switching the driving right to the human-machine co-driving control mode; When the risk danger value is within the third preset range, switch the driving right to the full control mode of the intelligent navigation system; The preset human-machine co-driving strategy based on the Nash game theory includes: Establish a comprehensive cost function with the driving right allocation coefficients of the driver and the intelligent driving system as independent variables and the sum of the ship operation amount and the error between the current navigation trajectory and the planned trajectory as the dependent variable; With the goal of minimizing the comprehensive cost function, determine the optimal ship control sequence; Based on the Nash game theory, solve the driving right allocation coefficients of the driver and the intelligent driving system corresponding to the optimal ship control sequence; The comprehensive cost function is: Wherein, is the number of update iterations, is the comprehensive cost function, is the error variable of the driver, is the error variable of the intelligent navigation system, and are parameters related to the driving right allocation, and are dynamic adjustment coefficients.

7. An intelligent control system for human-machine co-driving on the inland river water surface, characterized in that, Including a human-machine control driving right switching device for an inland water surface ship as described in claim 6.

8. An inland water surface ship, characterized in that, Including an inland water surface human-machine co-driving intelligent control system as described in claim 7.

Images