An NMPC parallelized controller for unmanned marine vehicles

Abstract

The present application relates to a kind of NMPC parallelization controller for unmanned marine vehicle.The three degrees of freedom kinematics and dynamics mathematical model of unmanned marine vehicle and NMPC control algorithm are built on the controller, and FPGA is used to parallelize the algorithm model and speed up operation.The controller designs network distributed architecture with Linux+MQTT structure, and according to this architecture, the underlying Linux operating system is customized and developed, and the unmanned marine vehicle integrated management system with GNC hardware management library and MQTT software communication server as the core is designed and developed on the underlying Linux operating system, finally the parallelization of NMPC controller algorithm with unmanned marine vehicle as controlled object is realized by using unmanned marine vehicle to deploy the controller, and the purpose of improving the control performance of unmanned marine vehicle by using FPGA to accelerate the algorithm and further improve the control performance of unmanned marine vehicle is achieved.

Description

Technical Field

[0001] This invention belongs to the technical field of control methods for unmanned marine vehicles, and particularly relates to an NMPC parallel controller for unmanned marine vehicles. Background Technology

[0002] Unmanned marine vehicles (UUVs) are characterized by low cost and high maneuverability, and have become one of the main means of marine exploration and surface operations. Facing the complex and unknown marine environment, the success of UUV operations depends on their intelligent control capabilities. Traditional control algorithms, such as PID control, often require resetting PID parameters under changing environmental conditions and altered state; otherwise, ideal control results are difficult to achieve. Although Non-Machine-Machine Control (NMPC) has significant advantages over PID control under complex models and uncertain conditions, its complex algorithms, high computational load, and poor real-time performance have limited its application to simulation and research. Finding the optimal solution quickly within a limited sampling time is a major challenge that urgently needs to be addressed to expand the application areas of NMPC.

[0003] Currently, the most widely used heterogeneous hardware acceleration platforms for accelerating control algorithms are graphics processing units (GPUs), application-specific integrated circuits (ASICs), and field-programmable gate arrays (FPGAs). GPUs suffer from high power consumption, poor flexibility, and poor real-time performance; while ASICs have low power consumption and fast computing speed, their design cycles are long and their flexibility is weak; and research on using FPGAs for parallelizing NMPC algorithms in unmanned marine vehicles is still in its infancy. Summary of the Invention

[0004] To overcome the aforementioned shortcomings of existing unmanned marine vehicle (UAV) control algorithms, this invention proposes a novel parallelized NMPC controller for UAVs. This invention achieves parallelization of the NMPC controller algorithm with the UAV as the controlled object by deploying the NMPC controller on the UAV, and simultaneously utilizes FPGA for hardware acceleration of the algorithm, thereby improving the control performance of the UAV.

[0005] This invention is achieved through the following technical solution:

[0006] This invention provides an NMPC parallel controller for unmanned marine vehicles, comprising a controller hardware system and a controller software system; wherein:

[0007] The controller hardware system includes: (1) an FPGA algorithm accelerator module based on NMPC; (2) a BMS power management module; (3) a high-speed information transceiver parsing and analog input / output module; and (4) an onboard core controller module.

[0008] The controller software system includes: (1) a low-level Linux operating system based on PetaLinux; (2) a network distributed software integrated management system based on Linux; and (3) a task planning system based on Python.

[0009] Furthermore, this invention establishes a three-degree-of-freedom kinematic and dynamic mathematical model and an NMPC control algorithm for an unmanned marine vehicle on an NMPC parallel controller, and uses an FPGA to perform parallel acceleration computation on the algorithm model; wherein:

[0010] The kinematic model is as follows:

[0011] The dynamic model is as follows:

[0012] In the above formula, η = [xy] T ∈R 2 ψ is the position vector of the origin of the unmanned marine vehicle's coordinate system in the ground coordinate system; ψ is the heading angle of the unmanned marine vehicle, used to represent its attitude; v h The generalized velocity vector, v, is used to represent the position and state vector of the system. h =[uvr] T ∈R 3 , (u, v)∈R 2 Let be the velocities of the unmanned marine vehicle along the x and y axes in the carrier coordinate system, respectively; let r be the angular velocity of the unmanned marine vehicle rotating along the z-axis in the carrier coordinate system; τ is the generalized force vector, and τ = [XYN]. T ∈R 3 It is used to represent the force vectors acting on the system, namely the lateral and longitudinal control forces and the z-axis torque of the unmanned marine vehicle; M is the system inertia matrix, and M is symmetric positive definite, i.e., M = M T D(v) is the damping matrix, and it is positive definite, i.e., v T D(v)v>0,

[0013] The expressions for each matrix are as follows:

[0014]

[0015]

[0016]

[0017] In the above formula, v h τ is the input to the dynamic model; substituting M and D(v) yields the solution for the next time step. Then Integrating yields v for the next time step. h ; at this time v h The state information of the unmanned marine vehicle in the northeast-east coordinate system is obtained by inputting the kinematic model and substituting J(ψ) into it. Finally, the state information of the unmanned surface vessel in the geographic coordinate system is obtained by integrating the GPS latitude and longitude information of the unmanned surface vessel for correction.

[0018] Furthermore, the NMPC control algorithm built in the NMPC parallel controller for unmanned marine vehicles of this invention includes an NMPC path tracking control method, which includes the following process:

[0019] The spatial curves without time information to be tracked are converted into parametric expression path forms; then the kinematic and dynamic models of the system are transformed into a system of differential equations and adjusted according to the target spatial curves to obtain a mathematical motion model based on state deviations; a performance function for a constrained optimization problem is designed for the above model; finally, the optimization problem to be solved is transformed into a problem that minimizes the performance function; and the solution is performed using a PSO solver accelerated by the FPGA.

[0020] Furthermore, the performance function for designing constrained optimization problems for the above-mentioned model in the NMPC parallel controller of the present invention includes the terminal cost function for designing the performance function, the weight matrix of the state cost, the weight matrix of the input cost, and the weight matrix of the terminal cost.

[0021] Furthermore, this invention establishes a hardware-accelerated calculation method for trigonometric functions on the NMPC parallel controller for unmanned marine vehicles. The CORDIC algorithm is used to convert floating-point calculations of trigonometric functions into recursive calculations achieved through shifting and addition / subtraction operations.

[0022] Furthermore, in the NMPC parallel controller for unmanned marine vehicles of the present invention, the hardware acceleration calculation method for the trigonometric function is designed as a pipelined method and an iterative reuse method in the FPGA algorithm accelerator module based on NMPC.

[0023] Furthermore, in the NMPC parallel controller for unmanned marine vehicles of this invention, the underlying Linux operating system based on PetaLinux is customized using PetaLinux and is equipped with a network distributed architecture based on Linux+MQTT. It performs general management of all hardware devices through the GNC underlying hardware management library interface, and manages data using a "publish-subscribe" network distributed communication method.

[0024] Furthermore, in the NMPC parallel controller for unmanned marine vehicles of the present invention, the FPGA algorithm accelerator module based on NMPC includes: an NMPC controller module, a universal asynchronous transceiver serial port module, and a clock phase-locked loop (PLL) module.

[0025] The NMPC controller module includes: a matrix addition / multiplication module, a trigonometric function operation CORDIC module, an objective function calculation module, a PSO particle swarm solver module, and a random number generator RNG module.

[0026] Furthermore, in the NMPC parallel controller for unmanned marine vehicles of the present invention, the Linux-based network distributed software integrated management system adopts the ARM-based GNC hardware management library and MQTT network distributed structure. By designing the GNC encapsulation library, the operation interface of the FPGA hardware platform is encapsulated, so that the controller system can directly call the FPGA hardware interface through the software API of the GNC encapsulation library without further data conversion.

[0027] An MQTT server is set up on a Linux system to implement a network distributed topology architecture for communication messages. The MQTT server is used as a communication middleware. All sensor data, system status, and temporary data of program operation are stored in the communication middleware through publication. Any system and program can obtain real-time data and status through subscription.

[0028] Furthermore, in the NMPC parallel controller for unmanned marine vehicles of the present invention, the Python-based task planning system includes: a task planning and timing control subsystem, a navigation control calculation subsystem, a BMS monitoring and management subsystem, a security subsystem, a communication and sensing subsystem, and an execution control subsystem.

[0029] In summary, the NMPC parallel controller of this invention for unmanned marine vehicles has the following characteristics:

[0030] (1) This invention takes unmanned marine vehicles as the controlled object and realizes the parallelization of the NMPC control algorithm, which significantly improves the control performance of unmanned marine vehicles.

[0031] (2) This invention is the first to use FPGA to perform parallel hardware acceleration of NMPC control algorithm for unmanned marine vehicles. On the one hand, FPGA can achieve a higher degree of parallelism through logic concurrency and task pipeline. On the other hand, due to the customizability of FPGA hardware resources and the reconfigurability of internal logic, it can run tasks directly on hardware circuits, thereby greatly speeding up the running speed and saving internal resource space. Attached Figure Description

[0032] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the following drawings are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 Flowchart for equipping unmanned marine vehicles with NMPC parallel controllers to perform mission tasks;

[0034] Figure 2 A schematic diagram of a three-degree-of-freedom kinematics and dynamics model based on an unmanned marine vehicle;

[0035] Figure 3 This is a schematic diagram of the controller system structure;

[0036] Figure 4 This is a schematic diagram of the FPGA hardware acceleration module architecture;

[0037] Figure 5 This is a schematic diagram of the CORDIC algorithm.

[0038] Figure 6 A schematic diagram of the overall design framework of the controller software system;

[0039] Figure 7 Flowchart for designing a customized underlying Linux system;

[0040] Figure 8 This is a schematic diagram of the overall architecture of the integrated management system. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. This invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this invention.

[0042] At the same time, it should be understood that the scope of protection of the present invention is not limited to the specific embodiments described below; it should also be understood that the terminology used in the embodiments of the present invention is for describing specific embodiments and not for limiting the scope of protection of the present invention.

[0043] Example: An NMPC Parallel Controller for Unmanned Marine Vehicles

[0044] Figure 1The flowchart illustrates the mission execution process for an unmanned marine vehicle equipped with an NMPC parallel controller. The unmanned marine vehicle performs its mission in the designated sea area. After system initialization, the NMPC controller parses and integrates real-time monitoring data from various sensors and awaits mission instructions from the shore-based system. Upon receiving mission instructions, it calculates the target waypoint location information and plans its motion path based on its own position information. Finally, it controls the underlying motion execution controller to track the waypoint path until the mission is completed. The shore-based monitoring and control platform is responsible for issuing mission instructions and monitoring the surface vessel's real-time status until the mission concludes.

[0045] Figure 2 This is a three-degree-of-freedom kinematic and dynamic model based on an unmanned marine vehicle. The kinematic model is as follows: The dynamic model is as follows:

[0046] In the above formula, η = [xy] T ∈R 2 ψ is the position vector of the origin of the unmanned marine vehicle's coordinate system in the ground coordinate system; ψ is the heading angle of the unmanned marine vehicle, used to represent its attitude; v h It is a generalized velocity vector used to represent the position and state vector of the system, where v h =[uvr] T ∈R 3 , (u, v)∈R 2 Let be the velocities of the unmanned marine vehicle along the x and y axes in the carrier coordinate system, respectively; r be the angular velocity of the unmanned marine vehicle rotating along the z-axis in the carrier coordinate system; τ be the generalized force vector, and τ = [XYN]. T ∈R 3 It is used to represent the force vectors acting on the system, namely the lateral and longitudinal control forces and the z-axis torque of the unmanned marine vehicle; M is the system inertia matrix, and M is symmetric positive definite, i.e., M = M T D(v) is the damping matrix, and it is positive definite, i.e., v T D(v)v>0, The specific expressions for each matrix are as follows:

[0047]

[0048]

[0049]

[0050] In the above formula, v h τ is the input to the dynamic model; substituting M and D(v) yields the solution for the next time step. Then Integrating yields v for the next time step.h . v at this time h The kinematic model input is used to calculate J(ψ) to obtain the state information of the unmanned marine vehicle in the northeast-northeast coordinate system at the next moment. Finally, by fusing the GPS latitude and longitude information of the unmanned surface vessel for correction, its state information in the geographic coordinate system can be obtained.

[0051] Figure 3 This is a schematic diagram of the controller system architecture. The controller consists of six structural modules: FPGA underlying hardware subsystem, sensor monitoring subsystem, execution controller subsystem, wireless communication subsystem, ARM integrated management subsystem, and shore-based operation platform.

[0052] The FPGA underlying hardware subsystem is further divided into an NMPC algorithm parallel acceleration module (i.e., an NMPC-based FPGA algorithm accelerator module), a high-speed information transceiver and parsing module, an analog control input / output module (i.e., a high-speed information transceiver and parsing and analog input / output module), a BMS power management module (i.e., a BMS power management module), and an onboard core controller module. This subsystem is mainly responsible for managing all underlying hardware devices, including parallel acceleration of the NMPC algorithm, acquisition and parsing of internal and external information, communication with the external environment, high-precision and fast input / output at the underlying level, and comprehensive power management.

[0053] The execution controller subsystem manages and controls the corresponding execution controllers, such as servos and thrusters, by receiving control commands issued by the FPGA underlying hardware subsystem. The wireless communication subsystem relies on WiFi modules, Iridium satellite communication modules, and radios to achieve wireless communication between the unmanned marine vehicle and the shore-based operating platform.

[0054] The ARM integrated management subsystem serves as the intelligent brain of the unmanned marine vehicle, primarily responsible for its overall mission management and decision-making. The shore-based operations platform, on the other hand, is the remote human-machine interface platform for the unmanned marine vehicle, possessing functions such as status monitoring and display, basic motion control, mission issuance and modification, and emergency stop and recovery.

[0055] Figure 4 This is a schematic diagram of the FPGA hardware acceleration module architecture. It specifically includes: an NMPC controller module, a universal asynchronous transceiver serial port module, and a clock phase-locked loop (PLL) module. The NMPC controller module specifically includes a matrix addition / multiplication operation module, a trigonometric function operation CORDIC module, an objective function calculation module, a PSO particle swarm optimization solver module, and a random number generator (RNG) module.

[0056] Figure 5This is a diagram illustrating the principle of the CORDIC algorithm. The CORDIC algorithm is a recursive calculation method that obtains trigonometric function values ​​through shifting and addition / subtraction operations. To solve for the phase of a point on a coordinate axis, the point needs to be rotated. When the point is rotated to the point where it first coincides with the x-axis, the phase magnitude after rotation is the phase of the point. The principle of the CORDIC algorithm is to transform this rotation process into a "discrete" rotation, rotating a specific angle each time, for N rotations. During the rotation, it continuously checks whether the point falls on the x-axis. If it falls above the x-axis, it rotates counterclockwise (increasing the phase); if it falls below the x-axis, it rotates clockwise (decreasing the phase). The phase changes during the rotation are accumulated, and finally, the phase of a point can be obtained. The specific algorithm is as follows:

[0057] First, define the rotation angle θ in trigonometric functions, and then decompose θ into several rotations. Each rotation angle satisfies tanθ. i =2 -i , where i is the number of rotations. Then, for any point P in the coordinate system... tar Let the coordinates be (x, y), and its phase be θ. Suppose that after the final rotation, it lands on point P0 on the x-axis, with coordinates (x0, y0). Then the angle between P and P0 is θ, so let x0 = 1 and y0 = 0. Then we can obtain:

[0058]

[0059]

[0060] Then θ can be equivalent to a series of directions d i The angle is θ i The coordinate rotation transformation. Where, d i Clockwise is 1, counterclockwise is -1. That is:

[0061]

[0062] Combining the above equations, we can obtain:

[0063]

[0064] in,

[0065] And because when a number is equal to 2 -i Multiplication is equivalent to right-shifting this number by i bits in binary form, therefore let tan(d) i θ i )=d i 2 -i Therefore, after n iterations, we can obtain approximate values ​​for sin and cos:

[0066]

[0067] in,

[0068] Figure 6 This is a schematic diagram of the overall design framework of the controller software system. The controller system software uses a customized version of Linux based on PetaLinux as the underlying operating system; the middle layer is a network-distributed software integrated management system architecture based on the GNC library and MQTT server, designed and developed on the Linux system; the top-level task planning system is designed using Python, and it specifically includes a task planning and timing control subsystem, a navigation control calculation subsystem, a BMS monitoring and management subsystem, a security subsystem, a communication and sensing subsystem, and an execution control subsystem.

[0069] Figure 7 The flowchart for designing a customized underlying Linux system includes the following steps:

[0070] Step 1: Create a Vivado project and generate the .xsa system file;

[0071] Step 2: Configure the Petalinux environment and create a Petalinux project;

[0072] Step 3: Configure the Petalinux project custom system;

[0073] Step 4: Configure the Linux kernel;

[0074] Step 5: Configure the root system file;

[0075] Step 6: Configure the device tree file;

[0076] Step 7: Compile the Petalinux project and create the .BIN boot file;

[0077] Step 8: Create an SD boot card.

[0078] Figure 8This is a schematic diagram of the overall architecture of the integrated management system. The software integrated management system design adopts an ARM-based GNC hardware management library and an MQTT network distributed structure. The control system design develops a GNC encapsulation library to implement the FPGA hardware platform operation interface, allowing the controller system to directly call the FPGA hardware interface through the software API of the GNC encapsulation library without further data conversion. Secondly, a Message Queuing Telemetry Transport (MQTT) server is built on a Linux system to implement a network distributed topology architecture for communication messages. Using the MQTT server as a communication middleware, all sensor data, system status, and temporary program execution data can be stored in the communication middleware through publishing. Any system or program can obtain real-time data and status by subscribing.

[0079] The shore-based monitoring and control system remotely connects to the NMPC controller system via WiFi wireless communication technology to achieve human-machine interaction. This allows for data acquisition, real-time monitoring, remote control, and emergency power-off procedures via the shore-based operation platform. The security system determines whether the unmanned marine vehicle is experiencing abnormal conditions such as water leakage, overpressure, or overcurrent by analyzing data from leakage sensors and voltage / current data. In the event of abnormal operation, it will immediately cut off the power to the unmanned marine vehicle to prevent damage to critical equipment due to short circuits or fires.

[0080] The navigation and control calculation system issues control commands to the underlying execution controller based on the control results provided by the NMPC-based FPGA algorithm module. This controls the basic direction and speed of the unmanned marine vehicle (UAV), while simultaneously recording the calculation results and the UAV's real-time path. The mission planning system, a top-level planning and control system, is responsible for the overall mission management and decision-making of the UAV. It monitors and records the UAV's status, and issues and modifies missions. It supervises and controls the UAV to complete its predetermined tasks according to a time-series flowchart.

[0081] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to obtain equivalent embodiments without departing from the scope of the technical solution of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the technical solution of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A parallel controller for NMPC (Network Multiprocessor Control) for unmanned marine vehicles, characterized in that, The controller includes a controller hardware system and a controller software system; wherein: The controller hardware system includes: (1) an FPGA algorithm accelerator module based on NMPC; (2) a BMS power management module; (3) a high-speed information transceiver parsing and analog input / output module; and (4) an onboard core controller module. The controller software system includes: (1) a low-level Linux operating system based on PetaLinux; (2) a network distributed software integrated management system based on Linux; and (3) a task planning system based on Python.

2. The NMPC parallel controller for unmanned marine vehicles according to claim 1, characterized in that, The controller is equipped with a three-degree-of-freedom kinematic and dynamic mathematical model of the unmanned marine vehicle and an NMPC control algorithm, and an FPGA is used to parallelize and accelerate the computation of the algorithm model; wherein: The kinematic model is as follows: The dynamic model is as follows: In the above formula, η = [xy] T ∈R 2 ψ is the position vector of the origin of the unmanned marine vehicle's coordinate system in the ground coordinate system; ψ is the heading angle of the unmanned marine vehicle, used to represent its attitude; v h The generalized velocity vector, v, is used to represent the position and state vector of the system. h =[uvr] T ∈R 3 , (u, v)∈R 2 Let be the velocities of the unmanned marine vehicle along the x and y axes in the carrier coordinate system, respectively; let r be the angular velocity of the unmanned marine vehicle rotating along the z-axis in the carrier coordinate system; τ is the generalized force vector, and τ = [XYN]. T ∈R 3 It is used to represent the force vectors acting on the system, namely the lateral and longitudinal control forces and the z-axis torque of the unmanned marine vehicle; M is the system inertia matrix, and M is symmetric positive definite, i.e., M = M T D(v) is the damping matrix, and it is positive definite, i.e. The expressions for each matrix are as follows: In the above formula, v h τ is the input to the dynamic model; substituting M and D(v) yields the solution for the next time step. Then Integrating yields v for the next time step. h ; at this time v h The state information of the unmanned marine vehicle in the northeast-east coordinate system is obtained by inputting the kinematic model and substituting J(ψ) into it. Finally, the state information of the unmanned surface vessel in the geographic coordinate system is obtained by integrating the GPS latitude and longitude information of the unmanned surface vessel for correction.

3. The NMPC parallel controller for unmanned marine vehicles according to claim 2, characterized in that, The NMPC control algorithm built in the controller includes an NMPC path tracking control method, which includes the following process: The spatial curves without time information to be tracked are converted into parametric expression path forms; then the kinematic and dynamic models of the system are transformed into a system of differential equations and adjusted according to the target spatial curves to obtain a mathematical motion model based on state deviations; a performance function for a constrained optimization problem is designed for the above model; finally, the optimization problem to be solved is transformed into a problem that minimizes the performance function; and the solution is performed using a PSO solver accelerated by the FPGA.

4. The NMPC parallel controller for unmanned marine vehicles according to claim 3, characterized in that, The performance function designed for the constrained optimization problem of the above model includes the terminal cost function, the weight matrix of the state cost, the weight matrix of the input cost, and the weight matrix of the terminal cost.

5. The NMPC parallel controller for unmanned marine vehicles according to claim 1, characterized in that, The controller is equipped with a hardware-accelerated calculation method for trigonometric functions. The CORDIC algorithm is used to convert floating-point calculations of trigonometric functions into recursive calculations achieved through shifting and addition / subtraction operations.

6. The NMPC parallel controller for unmanned marine vehicles according to claim 5, characterized in that, In the NMPC-based FPGA algorithm accelerator module, the aforementioned hardware acceleration calculation method for trigonometric functions is designed as a pipelined approach and an iterative reuse approach.

7. The NMPC parallel controller for unmanned marine vehicles according to claim 1, characterized in that, The underlying Linux operating system based on PetaLinux is customized using PetaLinux and is equipped with a network distributed architecture based on Linux+MQTT. It manages all hardware devices through the GNC underlying hardware management library interface and manages data using a "publish-subscribe" network distributed communication method.

8. The NMPC parallel controller for unmanned marine vehicles according to claim 1, characterized in that, The NMPC-based FPGA algorithm accelerator module includes: an NMPC controller module, a universal asynchronous transceiver serial port module, and a clock phase-locked loop (PLL) module. The NMPC controller module includes: a matrix addition / multiplication module, a trigonometric function operation CORDIC module, an objective function calculation module, a PSO particle swarm solver module, and a random number generator RNG module.

9. The NMPC parallel controller for unmanned marine vehicles according to claim 1, characterized in that, The Linux-based network distributed software integrated management system adopts an ARM-based GNC hardware management library and an MQTT network distributed structure. By designing a GNC encapsulation library, the operation interface of the FPGA hardware platform is encapsulated, so that the controller system can directly call the FPGA hardware interface through the software API of the GNC encapsulation library without further data conversion. An MQTT server is set up on a Linux system to implement a network distributed topology architecture for communication messages. The MQTT server is used as a communication middleware. All sensor data, system status, and temporary data of program operation are stored in the communication middleware through publication. Any system and program can obtain real-time data and status through subscription.

10. The NMPC parallel controller for unmanned marine vehicles according to claim 1, characterized in that, The Python-based task planning system includes: a task planning and timing control subsystem, a navigation control calculation subsystem, a BMS monitoring and management subsystem, a security subsystem, a communication and sensing subsystem, and an execution control subsystem.

Images