I. Introduction The number of electronic components and equipment installed on automobiles is constantly increasing, such as power seats, power windows and doors, operable roofs, adjustable steering wheels, and air conditioning systems; as well as various lights, windshield wipers, power door locks, defrosters, rearview mirrors, horns, various indicator lights, and various digital instruments (tachometers, speedometers, water temperature gauges, fuel gauges), etc. The number of automotive electronic control systems, sensors, actuators, and wires is also constantly increasing. The function of the automotive body control system is to achieve convenient and flexible integrated control of various components on the vehicle. In traditional body control systems, wiring harnesses are used to connect and directly control various electronic components through point-to-point connections. However, with the increase in components, the wiring harnesses inside the car are becoming increasingly complex, the usable space inside the car is becoming smaller, and due to the complexity of the wiring, the failure rate is increasing, resulting in higher manufacturing costs, increased design and maintenance difficulties, and significantly reduced reliability. How to modify the body control system to solve the above-mentioned problems has attracted increasing attention from automotive professionals both domestically and internationally. II. Bus-Based Body Control System Traditional body control systems use wiring harnesses not only to transmit signals but also to implement control logic between various devices through the harnesses, relays, and switch contacts. Since these devices are distributed throughout the body, a distributed control system is more suitable for body control systems. The new body control system uses a single bus to replace the complex point-to-point wiring harnesses, introducing intelligent control nodes that combine hardware and software to build the system. This involves connecting various devices to multiple intelligent control nodes distributed throughout the body. Each intelligent control node is an embedded processing unit with certain computing and storage resources. These nodes are connected via a bus, and the software within each node enables comprehensive control of the various devices. In other words, software logic replaces the hardware logic of traditional body control systems, resulting in greater flexibility and maintainability. CAN is a bus technology widely used in automobiles. Using CAN bus technology to build a network platform for the vehicle body control system replaces parallel wiring harnesses with a serial bus structure, achieving distributed multiplexing and facilitating information exchange and sharing between components. It also integrates real-time diagnostics, testing, and fault alarm functions, and can directly display fault locations through an information screen for easy maintenance. Functions can be added or removed without affecting other parts of the system. Using CAN bus technology to build a network platform for the vehicle body control system is the future development direction. However, how to design and develop the vehicle body control system software, establish user-friendly and standardized modeling and design methods, and corresponding development platforms are key issues that need to be addressed. III. Automata Model Modeling and Analysis The state of the vehicle body control system is reflected in the states of various devices. Changes in device states are driven by discrete events triggered by user operations, sensor detection, etc., leading to the dynamic evolution of the system state. The vehicle body control system is a typical discrete event control system, and is usually modeled using a finite automata model. A typical finite automaton is represented by a quintuple A = (S, E, η, y0, Sm) (1), where S is the set of states, E is the set of events, η is the state transition function, y0 is the initial state, and Sm is the set of final states. S is a non-empty set, y0 ∈ S, Sm ∈ S, and η: S × E → S. This means that if e ∈ E, s1 ∈ S, s2 ∈ S, when event e occurs, the system state changes from s1 to s2, and η maps the product of S and E to S. To model a vehicle body control system using a finite automaton model, we first need to determine the system's S, and then give the system's η. The vehicle body control system involves many devices, and the number of states of each device is also large. If we directly model the entire system, the system's state space S will be enormous. Assuming there are 20 devices, each device has 3 states, and the state of the vehicle body control system is determined by the states of all devices, then the system's state is a combination of all device states, corresponding to 320 states, making the state space extremely large. Furthermore, considering that η can be represented by a state transition matrix, state transition table, or state transition diagram, these three are equivalent and can be converted to each other. Taking the state transition matrix as an example, using rows to represent the current state of the state machine, columns to represent the next state to be reached, and the intersection of rows and columns to represent triggering events, we obtain a 320×320 matrix, resulting in a much larger state space. From the above analysis, it can be seen that using a finite automaton model to model the vehicle body control system results in the complexity of state combinations. In addition, using a finite automaton model, at most one operation can be performed at any given time in each state of the system, meaning it can only describe sequential systems and lacks the ability to describe concurrent events, but the vehicle body control system contains a large number of concurrent events and concurrent behaviors. To address the problems of using a finite automaton model to model the vehicle body control system, the authors propose a new modeling and design method for the vehicle body control system—the rule-based description method. The rule-based description method introduces a hierarchical modeling mechanism, decomposing the objects that make up the system into multiple layers, establishing a tree-like hierarchical model of the system objects, using logical rule expressions to describe the logical control relationships between system objects, and using messages to transmit control relationships. The system's control tasks are divided into multiple subtasks, distributed across the various layers of objects that make up the system, effectively reducing the complexity of system control. A message mechanism facilitates the handling of concurrent events and behaviors. IV. Layered Modeling Mechanism A rule-based description method is used to model the vehicle body control system. To reduce the complexity of system design, the system objects are decomposed into two layers: components and interfaces. The system objects are divided into multiple subspaces according to their compositional relationships. The system partitioning follows the principle of "high cohesion, low coupling," effectively reducing control complexity. The system's control tasks are divided into multiple subtasks, distributed across the various layers of objects that make up the system. High-level objects act as managers, coordinating the control tasks between the various components of the system; low-level objects include sensors and actuators, which directly interact with the outside world. Sensors perceive environmental information in real time and deliver it to high-level objects, while actuators mainly translate controller commands into actual physical actions and apply them to the environment. Logical control relationships exist between the objects at different layers of the system. These logical control relationships are transmitted through messages: notification messages are sent from low-level to high-level objects, and control (command) messages are sent from high-level to low-level objects. The vehicle body control system consists of multiple components, each containing one or more interfaces, forming a tree-like hierarchical model as shown in Figure 1. A component is a logical abstraction of each functionally independent device in the system, while an interface is an abstraction of the I/O ports of the control unit. For example, a car's headlights consist of two lights, left and right, which always illuminate simultaneously under normal conditions; logically, this can be defined as a "headlight" component. Headlights also have normal states such as high beam, low beam, and on/off, and fault states such as open circuit and short circuit. Logical control relationships exist between components and interfaces, including relationships between components, between components and interfaces, and between interfaces. These logical relationships are described using formalized logical rule expressions. V. Rule Description of Logical Control Relationships The state of the vehicle body control system is determined by the set of states of all components and interfaces that make up the system. Events trigger changes in the system state, which in turn cause changes in the states of components and interfaces. How these changes occur and the process of change are determined by the logical control relationships between components and interfaces. Logical control relationships are described by logical rule expressions. Changes in system state are manifested as logical control relationships being transmitted between components and interfaces via messages (when the two components involved in the logical control relationship are located in different control units, the control messages are transmitted via CAN), triggering state changes in the corresponding components and interfaces. Logical rule expressions, or simply rule expressions, are formal representations of the logical relationships between components and interfaces. Logical rule expressions can be seen as simplifications of ECA rules. An ECA rule can be expressed as follows: where E, C, and A are the event, condition, and action of the rule, respectively; P is an additional property describing the rule's behavior or state. The function of an ECA rule is: when a rule event occurs, the system checks the rule's condition in real time or at a specified time; if the condition is met, the rule's action is executed. Events in the body control system are triggered by user operation or sensor detection, causing changes in the state of corresponding devices. Therefore, events can also be treated as conditions in the expression, thus simplifying the expression. Using the Backus-Naur normal form syntax, the formal definition of logical rule expressions is given as follows. Definition 1 (Logical Rule Expression) A logical rule expression is defined as follows: := Left-hand → Right-hand Left-hand ::= Factor | Factor & Left-hand Right-hand ::= Factor | Factor & Right-hand The general form of a logical rule expression is: Factor & Factor & ... & Factor → Factor & Factor & ... & In a factor expression, the part to the left of the symbol "→" is called the left-hand component, and the part to the right is called the right-hand component. Both the left and right components consist of factors. When there is more than one factor, they are connected by "&" to represent "logical AND". The left-hand factor is the condition factor, and the right-hand factor is the response factor. Definition 1 gives the syntactic form of a logical rule expression, with the semantic meaning: if the left-hand component is true, that is, all condition factors in the left-hand component are true, i.e., the condition is met, then the right-hand component is executed, i.e., each response factor is executed. Using the Backus-Naur normal form syntax, the formal definition of the factor is given below. Definition 2 (Factor) Factor ::= (Factor Name = Factor Value) A factor is the basic building block of a logical rule expression. A factor consists of two parts: a factor name and a factor value, representing the component/interface and its state value, respectively. The control behavior in a vehicle body control system and the logical control relationships between components can be conveniently described using logical rule expressions. For example: For the following logical control relationship, if the dimmer switch is in the "low beam" position, the headlight switch is in the "headlight" position, and the ignition switch is in the "ON" state, then the car's headlights will illuminate as low beam. This logical control relationship can be formally represented as the rule (dimmer switch = low beam) & (headlight switch = headlight) & (ignition switch = ON) → (headlight = low beam). [align=center] Figure 2 Rule processing process of the rule-based description method[/align] The key to modeling the vehicle control system using the rule-based description method is to use logical rule expressions that approximate natural language to describe the logical control relationships between the various components of the system, forming a rule base. After the rule base is encoded, it is interpreted and executed by the system's control unit. The final system control process is the process of event triggering, rule matching, rule execution, and message passing, as shown in Figure 2. To modify the system's control logic, it is only necessary to modify the logical rule expressions in the rule base. The characteristics of the rule-based description method can be summarized as follows: 1. Naturalness: Similar to human thinking, intuitive, natural, easy to understand and explain, and convenient for human-machine information exchange. 2. Modularity: Rules are the most basic units in the rule base. Rules are independent of each other, do not call or modify each other, and are easy to add, delete, expand, and improve. 3. Clarity: Rules have a fixed format. Each rule consists of two parts: a condition and an operation. This unified format is easy to design and control, and also easy to perform consistency and integrity checks. 4. Independence: The rule base representing the system's logical control relationships is relatively independent from the inference engine that executes the rules. This facilitates the management and maintenance of the rule base and also the design and implementation of the inference engine. VI. Conclusion 1. A modeling method for a vehicle body control system based on CAN bus technology and using a rule-based description method was developed. A corresponding experimental environment and simulation, debugging, and testing tools were constructed and successfully applied to a test vehicle. 2. Using a rule-based description method greatly simplifies the development and maintenance of vehicle body control software and hardware for different vehicle models and functions. 3. When the vehicle model changes or new functions need to be added, only the user description file composed of logical rule expressions needs to be modified. This eliminates the need to spend a lot of time rewriting the system software, improving reusability, shortening the development cycle of new vehicles, and reducing development costs.