Immersive off-road mode switching mode

Abstract

The invention discloses an immersive off-road mode switching mode, and relates to the field of vehicle electronic technology and intelligent cabin man-machine interaction, and the method comprises the steps: responding to a switching instruction, and controlling a virtual camera to move towards an off-road scene along a planned path while preloading resources. The system divides a control state according to a transition progress: in a pre-interaction state that resources are ready but transition is not completed, the off-road instrument assembly is controlled to fade in display and open input, a user pre-operation instruction is temporarily stored in an instruction buffer queue through de-duplication logic, and a physical signal is temporarily not issued through a vehicle bus interface; and when the transition time is over and a fully activated state is entered, the queue is locked, and an instruction is converted into a standard protocol data unit to be sent to a chassis electronic control unit. Through the staged state machine and instruction buffer logic, three-dimensional visual rendering and a bottom layer control time sequence are decoupled, and continuity of interactive operation and deterministic transmission of a control instruction are realized while scene switching smoothness is ensured.

Description

Technical Field

[0001] This application relates to the fields of vehicle electronics technology and intelligent cockpit human-machine interaction, and in particular to an immersive off-road mode switching method. Background Technology

[0002] With the development of intelligent cockpit technology, in-vehicle human-machine interaction systems are gradually evolving towards three-dimensionality and refinement. In vehicles with off-road capabilities, switching driving modes involves not only adjustments to chassis dynamics parameters but also changes in the human-machine interface. Existing off-road mode switching solutions typically employ discrete application call logic, meaning that in response to physical button or touch commands, the system directly jumps from the current main interface to a separate off-road function page or pop-up window. This discontinuous page switching method disrupts the user's visual perception path in the virtual space, lacks spatial logic for scene transitions, and can easily lead to a sudden visual shift and spatial disorientation for the driver during mode switching.

[0003] Existing 3D interface rendering is usually a static preset animation, which fails to establish a dynamic mapping relationship with the real-time physical state of the vehicle (such as changes in vehicle speed and body posture). This results in a disconnect between the visual feedback content and the actual driving experience, reducing the immersion and realism of human-computer interaction.

[0004] Furthermore, the resource loading and real-time rendering of high-fidelity 3D off-road scenes are typically accompanied by high data throughput and computational load. In existing system architectures, the visual rendering process and the underlying control command sending logic often lack effective timing decoupling and coordination mechanisms.

[0005] When the system performs background loading of large-scale scene resources or foreground transition rendering, if the user attempts to configure functions, input interaction may be blocked, command response may be delayed, or even lost. In the absence of command buffering and quality of service arbitration mechanisms, repeated operations performed by impatient users during transition waiting periods can easily cause redundant data streams to momentarily preempt vehicle bus bandwidth, thereby interfering with the real-time and deterministic transmission of core chassis control signals and posing a potential hazard to vehicle driving safety. Summary of the Invention

[0006] This invention provides an immersive off-road mode switching method, aiming to solve the problems of visual feedback being disconnected from the underlying control logic, discontinuous interactive experience, and chaotic instruction processing during the transition period in existing in-vehicle human-machine interaction systems during driving mode switching.

[0007] The technical solution adopted in this invention is as follows:

[0008] The method operates within an in-vehicle human-machine interface system, which includes a cockpit domain controller, a display screen, an input device, and a vehicle bus interface. The method primarily includes the following processes: In response to a mode switching command received via the input device, the cockpit domain controller initiates the preloading of off-road driving scene resources in the background and simultaneously controls a virtual camera to move along a planned path from the currently displayed 3D scene to the off-road driving scene within the 3D virtual environment.

[0009] During this process, the cockpit domain controller determines the resource readiness time threshold based on the loading progress of core resources in the off-road driving scenario and sets the total transition time. This method divides the switching process into different control stages based on the aforementioned time parameters: when the current transition time is between the resource readiness time threshold and the total transition time, the system enters a pre-interaction state. At this time, the control system displays the off-road core instrument components on the screen and opens the input interface. The system stores the pre-operation commands received during this period in the command buffer queue and temporarily does not send physical control signals through the vehicle bus interface; when the current transition time reaches or exceeds the total transition time, the system enters a fully active state, reads the pre-operation commands in the command buffer queue, and sends them to the vehicle chassis electronic control unit through the vehicle bus interface.

[0010] Furthermore, regarding path planning for the virtual camera in 3D space, this method constructs a smooth transition path by setting a first spatial control point vector to control the tangent direction of the starting segment of the path and a second spatial control point vector to control the tangent direction of the ending segment of the path. The cockpit domain controller calculates a normalized time parameter based on the ratio of the current transition time to the total transition time, and uses a cubic Bézier curve algorithm to perform weighted calculations on the initial coordinate vector of the virtual camera, the main viewpoint coordinate vector of the target off-road driving scene, the first spatial control point vector, and the second spatial control point vector according to the normalized time parameter, thereby obtaining the instantaneous position vector of the virtual camera and achieving a smooth transition of viewpoint.

[0011] Furthermore, to ensure system stability during resource loading, the system enters a pure guidance state when the current transition time is less than the resource readiness time threshold. In this state, the cockpit domain controller only renders environmental background changes caused by the movement of the virtual camera, hides the off-road core instrument components, and disables interactive event responses from input devices to prevent users from performing invalid operations when resources are not ready.

[0012] Furthermore, to optimize command transmission efficiency and prevent conflicts, a deduplication and update mechanism is adopted for command storage logic in the pre-interaction state. The cockpit domain controller parses the received pre-operation commands into command data packets containing functional domain identifiers and traverses the current command buffer queue. If a data packet with the same functional domain identifier as the newly generated command data packet is detected in the queue, an overwrite operation is performed, removing the old command data packet and inserting the new command data packet into the queue; if no command with the same identifier is found, the new command data packet is directly appended to the end of the queue.

[0013] Furthermore, regarding the display control of the off-road core instrument component, the cockpit domain controller calculates the progress parameter of the current transition time relative to the starting point of the pre-interaction state, and based on this progress parameter, uses a linear interpolation algorithm to dynamically increase the transparency channel value and scaling ratio value of the component, so that the off-road core instrument component presents a gradually fading visual effect as the virtual camera moves.

[0014] Furthermore, during the command transmission phase in the fully activated state, the cockpit domain controller locks the command buffer queue, calls the protocol conversion module to convert the command data packets in the queue into standard protocol data units, and writes them sequentially into the network transmission buffer according to the timestamp order of command generation. If the command buffer queue is detected to be empty, a preset default configuration command is generated and sent to the vehicle chassis electronic control unit to ensure that the vehicle chassis always receives a definite control signal.

[0015] Furthermore, to improve interaction response speed and reduce user-perceived latency, in the pre-interaction state, when a touch event targeting the core off-road instrument components is detected, the system directly initiates the visual response thread without waiting for confirmation signals from the vehicle chassis electronic control unit. This thread modifies the rendering material properties of the touched target control object, adjusts its self-illumination intensity, and performs local geometric deformation on the target control object, providing real-time visual feedback to the user's pre-operation.

[0016] Furthermore, to enhance the immersive experience of the virtual environment, the motion parameters of the virtual camera are correlated with the vehicle's physical state. The cockpit domain controller acquires the vehicle's real-time speed and calculates dynamic field-of-view parameters based on this speed, causing the field-of-view to increase with vehicle speed. Simultaneously, it acquires the vehicle's vertical acceleration signal, uses a noise generation function to generate a pseudo-random three-dimensional vector, and superimposes this pseudo-random three-dimensional vector as a vibration offset into the instantaneous position vector of the virtual camera to simulate the bumps of a real road surface.

[0017] Furthermore, the resource preloading step employs an incremental asynchronous loading strategy, performing data decompression and video memory upload in batches within the allowable rendering time of each frame, thus avoiding blocking the rendering thread. Simultaneously, based on the current camera position and line-of-sight direction, the loading position of the off-road driving scene in the world coordinate system is calculated. This loading position is the coordinate of the scene after extending a preset distance along the line-of-sight direction and superimposing a vertical offset vector.

[0018] Furthermore, to ensure the stability of vehicle bus communication, the system monitors the real-time throughput bandwidth utilization of the vehicle bus interface. When the bandwidth utilization exceeds a preset safety threshold, a service quality priority arbitration mechanism is triggered, suspending the transmission of non-critical media data streams and prioritizing the transmission of standard protocol data units in the instruction buffer queue to ensure the real-time transmission of core control instructions.

[0019] This invention decouples scene rendering transitions from underlying control signal interactions through a time-axis-based, phased state control logic. During visual transitions, a pre-interaction state and command buffer mechanism are introduced, ensuring smooth 3D scene switching while allowing users to pre-set operations. Furthermore, the command queue, in its fully active state, ensures the timing accuracy and safety of vehicle mode switching control.

[0020] The above solution achieves the following beneficial technical effects:

[0021] This application decouples the visual transition cycle of a 3D scene from the timing of the underlying chassis control signals by setting a pre-interactive state and an instruction buffer queue. During visual resource loading or camera movement, the system allows users to configure functions in advance and temporarily store instructions, which are then executed uniformly after the transition ends and the system enters a fully active state. This mechanism eliminates the blocking effect of long-duration visual rendering on real-time control response, ensuring the continuity of user interaction and preventing unexpected fluctuations in vehicle driving status caused by direct issuance of instructions in intermediate transition states, thus improving the safety of driving mode switching.

[0022] This application utilizes a cubic Bézier curve algorithm to plan the spatial motion path of a virtual camera and maps the vehicle's real-time speed and vertical acceleration signals to the camera's dynamic field of view parameters and positional vibration offset. This technical feature ensures a smooth and physically consistent transition of the virtual viewpoint between different rendering scenes, keeping visual feedback dynamically synchronized with the vehicle's actual physical posture. It solves the problems of visual abrupt changes and spatial disorientation caused by traditional hard-switching methods, and enhances the perceptual consistency between the human-computer interface and the vehicle's driving status.

[0023] This application employs a deduplication and overwrite mechanism based on functional domain identifiers in the instruction buffer queue, combined with a service quality priority arbitration strategy for the vehicle bus interface. This design filters out redundant data packets generated by repeated user operations before instruction transmission and prioritizes control instruction transmission by suspending non-critical media streams under high bus load scenarios. This effectively reduces the ineffective occupation of vehicle bus bandwidth by instantaneous concurrent data, ensuring determinism and low latency in the transmission of core control instructions for off-road mode switching. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the hardware connection and data interaction of the immersive off-road mode switching system provided in this embodiment of the invention.

[0025] Figure 2 This is the main control flowchart of the immersive off-road mode switching method provided in the embodiment of the present invention.

[0026] Figure 3 This is a flowchart of the instruction buffer and visual collaborative control for driving mode switching provided in an embodiment of the present invention.

[0027] Figure 4 This is a flowchart of the deduplication and transmission process of the instruction buffer queue provided in the embodiments of the present invention. Detailed Implementation

[0028] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] See attached document Figure 1 This invention provides an immersive off-road mode switching method, which operates within an in-vehicle human-machine interaction system.

[0030] The in-vehicle human-machine interface system includes a cockpit domain controller, a display screen, input devices, and a vehicle bus interface. The cockpit domain controller connects to the display screen via a video output interface for outputting 3D rendered images. The cockpit domain controller connects to the input devices, which include physical buttons on the center console or a microphone array within the vehicle, for receiving user commands. The cockpit domain controller connects to the vehicle chassis electronic control unit via the vehicle bus interface to acquire real-time vehicle status data and send control commands. The cockpit domain controller internally runs a 3D rendering engine, which is used to construct and render a 3D virtual environment encompassing urban driving and off-road driving scenarios.

[0031] The immersive off-road mode switching method provided by this invention specifically includes the following execution process:

[0032] The cockpit domain controller receives mode switching commands via input devices. These commands are generated by the user pressing physical buttons or issuing specific voice commands. Upon receiving the command, the cockpit domain controller locks the currently displayed 3D scene and preloads the geometric and texture resources of the off-road driving scene in the background of the 3D rendering engine.

[0033] The cockpit domain controller plans a continuous camera movement path in the 3D virtual environment based on the main viewpoint coordinates of the currently displayed 3D scene and the main viewpoint coordinates of the target off-road driving scene. The cockpit domain controller controls the virtual camera in the 3D rendering engine to move along the planned camera movement path, thereby presenting a smooth, one-shot transition animation on the display screen from the currently displayed 3D scene to the off-road driving scene.

[0034] During the planning of the camera's movement path, the cockpit domain controller uses a cubic Bézier curve algorithm to calculate the instantaneous position of the virtual camera in three-dimensional space. Let... This is the current transition time. Define a normalized time parameter for the preset total transition duration. This is the ratio of the current transition time to the total transition duration. The instantaneous position vector of the virtual camera in 3D space. Calculate using the following formula:

[0035] ;

[0036] in, Virtual camera in normalized time parameters The instantaneous position vector in three-dimensional space at time; The normalized time parameter is defined as the current transition time. Total transition time The ratio (i.e.) (Values ​​range [0, 1]) The initial coordinate vector of the virtual camera in the starting scene (urban driving scene); The main viewpoint coordinate vector of the virtual camera in the target off-road driving scene; The first spatial control point vector is used to control the tangent direction and curvature of the starting segment of the path; The second spatial control point vector is used to control the tangent direction and curvature of the path's ending segment.

[0037] As the virtual camera moves along its path, the cockpit domain controller performs phased information loading and interactive control based on a timeline. The cockpit domain controller divides the transition process into a pure guidance state, a pre-interaction state, and a fully active state.

[0038] The cockpit domain controller defines and monitors three key time states: First, the pure guidance state, corresponding to the time threshold between zero transition time and the completion of loading of off-road core instrument resources. In this state, the cockpit domain controller controls the display screen to only show the environmental background changes caused by the movement of the virtual camera, and hides the off-road function user interface components and disables interactive input to prevent users from accidentally touching resources that have not yet been loaded. Second, the pre-interaction state, corresponding to the time threshold between the completion of resource loading and the completion of loading of the virtual camera reaching the target point. In this state, the cockpit domain controller controls the off-road core instrument components to render and display in the center area of ​​the display screen, and opens the input interface of the off-road core instrument components to receive pre-operation commands from the user. The cockpit domain controller stores the received pre-operation commands in the command buffer queue and blocks the channel for sending physical control signals to the vehicle chassis electronic control unit. Third, the fully active state, corresponding to the moment when the transition time reaches or exceeds the total transition time. In this state, the cockpit domain controller determines whether there are pre-operation commands in the command buffer queue. If it exists, the most recent pre-operation command with the most recent timestamp is read and sent to the vehicle chassis electronic control unit; if it is empty, the default configuration command is sent.

[0039] Through the above process, the cockpit domain controller achieves seamless switching from urban driving scenarios to off-road driving scenarios in the 3D rendering engine, and responds to the user's configuration requirements in advance before the transition animation ends.

[0040] Step 1: Identification and Resource Preprocessing of Mode Switching Requests. The cockpit domain controller polls signals in real time through the input interface module. When a trigger signal is detected, scene state locking is performed, current virtual camera parameters are recorded, and a background asynchronous loading process is started to upload the terrain height map, texture map, and ambient occlusion map data of the off-road scene to the video memory.

[0041] Step 2: Construction of the transition trajectory based on vehicle dynamics parameters. The cockpit domain controller constructs the transition path and incorporates the vehicle's real-time driving speed to correct the dynamic field of view. The cockpit domain controller calculates the dynamic field of view parameter, which is based on the reference field of view angle and superimposed with an increment that is proportional to the vehicle's real-time driving speed and varies sinusoidally over time, thereby generating a visual impact effect that is positively correlated with vehicle speed during the transition.

[0042] Step 3: Time-driven phased rendering and logic control. The cockpit domain controller maintains a global timer. During the resource-unready phase, scene blending rendering is performed and interaction is disabled; during the resource-ready phase but before the transition is complete, functional controls are drawn, touch is enabled, and commands are serialized and stored in a queue.

[0043] Step 4: Transition Final State Confirmation and Command Queue Execution. Upon completion of the transition, the cockpit domain controller fixes the virtual camera's view and checks the command queue. If it is not empty, the latest command is sent; if empty, default parameters are sent, and finally, all interactive permissions are unlocked.

[0044] The cockpit domain controller employs an incremental asynchronous loading strategy. During the execution of the data loading thread, the cockpit domain controller strictly limits the time slice length for processing the data stream in each frame, performing data decompression and uploading only within the allowable range of the current frame's rendering time to avoid screen stuttering. The cockpit domain controller calculates the loading position of the off-road driving scene in the world coordinate system based on the current line of sight.

[0045] This location is based on the current camera position, extending along the direction determined by the observation rotation matrix by a preset virtual scene interval distance, and superimposed with a vertical offset vector to correct for terrain height differences. Simultaneously, the cockpit domain controller sets initial invisible attributes for off-road scene objects and pre-calculates spherical harmonic illumination coefficients and fog density parameters, establishing a linear interpolation mapping table of illumination parameters from the current scene to the off-road scene.

[0046] The cockpit domain controller performs real-time path calculation within the 3D rendering engine. In addition to the aforementioned position calculations, the cockpit domain controller employs a quaternion spherical linear interpolation algorithm to calculate the instantaneous attitude of the virtual camera. This algorithm, on a four-dimensional hypersphere, interpolates the attitude along a great circle path based on the angle between the initial and target attitude quaternions, ensuring uniform rotational angular velocity. To simulate physical inertia, the cockpit domain controller remaps the time parameters using a nonlinear easing function and processes the original time scale using a smoothing step function, giving the camera motion a slow-fast-slow physical acceleration / deceleration feel.

[0047] The cockpit domain controller synchronously drives multiple visual effects layers. For the initial scene, the controller dynamically calculates transparency based on time parameters, gradually transitioning it from opaque to fully transparent. For the off-road scene, the reverse transition from fully transparent to opaque is executed. Simultaneously, the controller manages fog effects based on a depth buffer. In the first half of the transition, the fog density linearly increases to its peak; in the second half, it decreases from the peak to the target scene's set value, creating a visual effect of passing through fog. Furthermore, the controller synchronizes the particle system's emission density and lifespan with the transition progress, enhancing the dynamic atmosphere.

[0048] The cockpit domain controller maps vehicle dynamics data to rendering parameters. For speed, it dynamically adjusts the field of view to simulate the tunnel effect. For steering wheel rotation angle, the cockpit domain controller maps it to the yaw and roll angle offsets of the virtual camera, causing the camera's line of sight to follow the movement.

[0049] For vehicle vertical acceleration signals or engine speed signals, the cockpit domain controller superimposes programmed vibration noise onto the virtual camera path. The cockpit domain controller uses either the Berlin noise algorithm or the simplex noise algorithm to generate continuous pseudo-random three-dimensional vectors and calculates the vibration offset vector according to the following formula. :

[0050] ;

[0051] in, for The vibration offset vector superimposed on the camera path at any given moment; The basic vibration amplitude coefficient determines the baseline strength of the vibration. The road surface feedback gain coefficient is used to adjust the weight of the influence of vertical acceleration on the vibration amplitude; Instantaneous value of the vehicle's vertical acceleration; Noise ( A three-dimensional noise generation function (such as Berlin noise or simplex noise) returns a pseudo-random vector; The noise sampling frequency determines the speed and rhythm of the vibration; Current time.

[0052] See attached document Figure 2 The cockpit domain controller maintains a high-precision global timer, which is reset and started upon detecting the rising edge of a mode switch command. The total transition time is defined as the entire time span from the start of the mode switch to its completion. The total transition time is a preset constant (e.g., 3000 milliseconds) stored in non-volatile memory. Its value is ergonomically calibrated to ensure that the visual transition is neither too fast and dizzying, nor too slow and sluggish.

[0053] The cockpit domain controller sets a dynamic critical time node on the timeline, namely the resource readiness time threshold. The resource readiness time threshold is not a fixed constant, but is calculated in real time by the cockpit domain controller's resource management subsystem based on the loading progress of core resources in the current off-road driving scenario. The cockpit domain controller monitors the video memory upload status of core instrument components in off-road mode (including the inclinometer texture, terrain selection knob model, and compass control). Once all the loading completion flags of the aforementioned core resources are set, the cockpit domain controller marks the current moment as the resource readiness time threshold. To ensure the stability of the logic, the cockpit domain controller is configured... The maximum value must be strictly less than the total transition time. .

[0054] Based on the current count value of the global timer The cockpit domain controller strictly divides the switching process into three logical phases and defines system state functions. as follows:

[0055] ;

[0056] in, time The system's logical state; The current count value of the global timer; Resource readiness time threshold, which is the moment when the core resources for off-road driving are fully loaded; Preset total transition duration (e.g., 3000 milliseconds); Pure guide mode (only render the background, disable interaction); Pre-interactive state (rendering UI, allowing input, blocking control); Fully activated (transition complete, command issued)

[0057] Phase 1: Pure Guiding Zone (

[0058] Within this range, the system is in a pure guidance state. The cockpit domain controller determines that the core functional components of the off-road mode are not yet ready in the graphics processing unit. Therefore, the cockpit domain controller only performs virtual camera displacement calculations and background environment rendering. The cockpit domain controller forcibly blocks touch interruption requests from the touchscreen and does not draw any user interface elements related to off-road functions on the interface to ensure that the user's visual focus is on the dynamic process of evacuating from the urban environment, while avoiding rendering errors due to resource shortages.

[0059] Phase Two: Pre-interaction Interval (

[0060] Within this range, the system is in a pre-interaction state. This is a crucial stage in achieving visual and interactive collaboration in this invention. At this point, although the virtual camera has not yet reached the main viewpoint of the target off-road scene (i.e., the transition animation is still in progress), the cockpit domain controller confirms that the core instrument components are ready for rendering. The cockpit domain controller overlays and renders the off-road mode operation interface on top of the 3D scene and opens the input event listening port for this interface. The cockpit domain controller allows users to pre-select terrain and drive modes via voice or touch at this stage, but at the logic layer, the cockpit domain controller intercepts all control messages attempting to be written to the vehicle controller and redirects them to the memory buffer.

[0061] Phase 3: Full Activation Zone

[0062] Within this range, the system is in a fully active state. At this point, the virtual camera has precisely positioned itself at the target coordinates in the off-road scene, and the transition animation ends. The cockpit domain controller fully unlocks the vehicle chassis control channels and begins periodically processing instructions in the memory buffer and subsequent real-time user commands. The cockpit domain controller switches the system logic from a transient state to a steady state, entering the normal off-road mode cyclic monitoring process.

[0063] The cockpit domain controller establishes a user interface component management list within the graphics rendering pipeline. In off-road mode, the cockpit domain controller categorizes all visual elements into two types of objects: a core set of interactive components and a set of auxiliary information components. The core set of interactive components includes a terrain mode selection knob, a drive mode switch, and core vehicle status gauges; the auxiliary information component set includes an altitude display control, a barometric pressure data control, and a compass orientation control. The cockpit domain controller executes differentiated lifecycle control logic for these two types of objects.

[0064] During the pure guidance state (i.e., the current transition time) Less than the resource ready time threshold The cockpit domain controller sets the rendering attributes of all objects in the core interaction component set and the auxiliary information component set to invisible, and disables the ray detection target attribute of all objects. At this time, the graphics processing unit does not perform fragment color viewing operations for the aforementioned components, and the input interface module does not perform collision detection on touch signals in the aforementioned component areas. When the system enters the pre-interaction state (i.e., the current transition time...), Greater than or equal to the resource readiness time threshold And less than the total transition time At this point, the cockpit domain controller triggers the instantiation and rendering process of the core interactive component set. The cockpit domain controller calculates the normalized progress parameters of the current moment relative to the start of the pre-interaction phase. And based on this parameter, dynamically update the transparency channel value of the core interactive components. and scaling ratio values

[0065] ;

[0066] in, The transparency channel value of the core interactive component at time t; Current transition time; Resource readiness time threshold; Preset fade-in transition duration; An interval constraint function ensures that the calculation results are strictly limited to a certain range. and Between (0 to 1).

[0067] When the system enters a fully active state (i.e., the current transition time) Greater than or equal to the total transition time When the system is activated, the cockpit domain controller updates the interaction response flags of the core interaction component set to the execution mode, connecting the signal path between the user interface events and the vehicle control bus. Simultaneously, the cockpit domain controller initiates the rendering process for the auxiliary information component set. Using a linear interpolation algorithm similar to that of the core interaction components, the cockpit domain controller controls the auxiliary information components to fade in from the screen edge or background. Therefore, all user interface components in off-road mode complete their full lifecycle evolution from dormancy to activation, forming a complete human-machine interface.

[0068] The state control model is established and executed by the cockpit domain controller through the following detailed technical means:

[0069] The cockpit domain controller constructs a deterministic finite state machine model at the system kernel layer to precisely coordinate the concurrent behavior among the visual rendering subsystem, the human-machine interface input subsystem, and the vehicle bus communication subsystem. The cockpit domain controller defines a time-varying three-dimensional state vector. , used to describe any time System control logic:

[0070] ;

[0071] in, at any time The three-dimensional system control state vector; The rendering visibility logic value of user interface components (0 for hidden, 1 for visible); Input the activation state logic value of the event listener (0 for sleep, 1 for active). The logic value for granting permission to send physical control signals (0 for blocking, 1 for enabling).

[0072] The cockpit domain controller uses the current global timer value. Compared with the preset resource readiness time threshold and total transition time Real-time comparison is performed, and the state vector is dynamically updated based on the comparison results. The components:

[0073] Logical blocking state:

[0074] When the cockpit domain controller determines At this time, the cockpit domain controller sets the state vector to [0, 0, 0]. In this state, the cockpit domain controller sets the rendering level of the human-machine interface to transparent and cancels the event callback functions of the input device. The cockpit domain controller ensures that no computing resources are allocated to processing user input at this time, thereby concentrating all computing power on the decompression and loading of background scene resources.

[0075] Logical buffer state:

[0076] When the cockpit domain controller determines At that time, the cockpit domain controller sets the state vector to

[0077] [1, 1, 0]. This is the core feature of the state control model.

[0078] The cockpit domain controller commands the graphics processing unit to begin drawing the core operating controls for off-road mode.

[0079] The cockpit domain controller activates the listening threads of the touchscreen and voice module, allowing the capture of the user's selection intent.

[0080] The cockpit domain controller activates a logic gate at the Controller Area Network (CAN) transmit interface. At this time, although the data packets generated by the input subsystem are valid, due to... If the value is 0, the cockpit domain controller redirects the routing target of the data packet from the network send buffer to the local memory instruction queue. This mechanism ensures that the vehicle chassis electronic control unit will not receive any instructions that could easily cause sudden changes in the vehicle's physical attitude (such as sudden suspension rise or fall) before the visual transition is complete, thus guaranteeing driving safety.

[0081] Logical conduction state:

[0082] When the cockpit domain controller determines At that time, the cockpit domain controller sets the state vector to .at this time, When the value is flipped to 1, the cockpit domain controller opens the transmission gate of the controller area network. The cockpit domain controller first scans the local memory instruction queue. If there are data packets that were backed up during the logical buffer state, the cockpit domain controller pushes these data packets into the network transmission buffer in timestamp order. Subsequently, the cockpit domain controller restores the data flow routing to direct pass-through mode, that is, subsequent user instructions will be directly converted into bus messages and sent immediately.

[0083] Through precise control of the aforementioned state vectors, the cockpit domain controller strictly defines the temporal relationship of the three actions of seeing, pointing, and moving on a microsecond-level time scale, and achieves stability of phased information loading and interactive pre-response at the underlying logic level.

[0084] See attached document Figure 3 The cockpit domain controller monitors touch interruption signals generated by the input devices in real time. When the cockpit domain controller is in the pre-interaction state (i.e., the current transition time)... At the resource ready time threshold Total transition time When a touch event is detected on the display screen, the cockpit domain controller immediately obtains the two-dimensional screen coordinates of the touch point. The cockpit domain controller uses the projection matrix and view matrix of the current virtual camera to emit a virtual ray from the touch point coordinates into the interior of the 3D virtual scene. The cockpit domain controller performs ray intersection detection on the geometric mesh colliders in the off-road mode function control set to determine the target control object that the user intends to select.

[0085] Once the cockpit domain controller confirms the selected target control object (such as the sand mode icon), it immediately branches off a separate visual response thread within the graphics rendering pipeline. This visual response thread does not wait for confirmation signals from the vehicle chassis electronic control unit; instead, it directly modifies the rendered material property instance of the target control object.

[0086] The cockpit domain controller uses physically based rendering (PBR) technology to dynamically adjust the emissivity and surface roughness properties of target control objects. The cockpit domain controller defines a material highlight evolution function. Used to control the visual dynamics of the target control object the moment it is selected:

[0087] ;

[0088] in, Since touch trigger The self-illuminating radiation brightness of the target control surface at any given time; The base brightness of the control when it is not selected; Peak brightness of the control when it is selected; Time elapsed since the touch event was triggered; Preset duration of a single highlight animation cycle; Pi (π)

[0089] Simultaneously, the cockpit domain controller performs a geometric transformation on the target control object to simulate the mechanical deformation of a physical button being pressed. The cockpit domain controller modifies the local transformation matrix of the target control object, making a small displacement along the opposite direction of the normal, and reducing the local scaling ratio to a preset percentage (e.g., 95%) of the original size. This geometric deformation occurs synchronously with material highlighting, providing the user with immediate tactile visual confirmation.

[0090] To clearly distinguish between the pre-selected and active states, the cockpit domain controller generates a specific particle halo around the target control object in the pre-interactive state. The cockpit domain controller instantiates a ring-shaped particle emitter and attaches it to the anchor point of the target control object. This emitter emits a semi-transparent, low-speed rotating stream of particles, and the particles' color is set to an intermediate color (e.g., amber) that differs from the normally active color. This visual feedback conveys a clear message to the user: the system has successfully captured the operation command and stored it in the buffer queue, awaiting the end of the transition animation to execute the final physical switch.

[0091] Through the aforementioned instantaneous material changes, geometric deformations, and particle effects, the cockpit domain controller eliminates the sense of lag that can easily occur when transition animations are not yet finished, ensuring that users receive a definite visual response the moment they initiate an operation.

[0092] The cockpit domain controller allocates a dedicated memory area in random access memory as an instruction buffer. The cockpit domain controller uses a hash-mapped linked list structure based on functional domain classification to construct the instruction buffer queue. This data structure aims to solve the state jitter problem that can easily occur when users trigger conflicting configuration instructions consecutively within a very short period.

[0093] The cockpit domain controller defines a standardized command data packet structure. Each command data packet contains the following key fields: a unique command identifier, a functional domain identifier, parameter payload, and a generation timestamp. The functional domain identifier defines the scope of the command, such as a terrain mode selection domain, a suspension height adjustment domain, or a differential lock control domain.

[0094] When the cockpit domain controller receives an operation command from the user in the pre-interactive state, the cockpit domain controller first parses the operation command into a command data packet to be queued. The cockpit domain controller executes the conflict detection and queue update algorithm. The cockpit domain controller traverses the current instruction buffer queue. Check if there is a stored instruction data packet. , making Functional domain identifiers and The functional domain identifiers are the same.

[0095] If the cockpit domain controller retrieves an old instruction data packet with the same functional domain identifier in the instruction buffer queue The cockpit domain controller performs an overwrite operation. The cockpit domain controller will then overwrite the old command data packets. Remove from the queue and add the new instruction data packet. Insert at the end of the queue. If the cockpit domain controller does not find an instruction with the same functional domain identifier, then directly insert it at the end of the queue. The instruction is appended to the end of the queue. The cockpit domain controller strictly defines the queue update process through set operation logic: that is, it removes all old instructions with the same function domain identifier as the new instruction from the current queue set, and then adds the new instruction to the set, thus forming the instruction buffer queue set for the next moment.

[0096] In addition, the cockpit domain controller sets a capacity safety threshold for the command buffer queue. The cockpit domain controller monitors the total number of commands in the queue in real time. If the total number of commands exceeds the preset depth limit, the cockpit domain controller triggers an overflow protection mechanism. The overflow protection mechanism discards commands from non-critical functional domains based on timestamps, prioritizing the retention of core control commands related to the powertrain and chassis systems.

[0097] Through the aforementioned buffer logic based on functional domain overriding, the cockpit domain controller ensures that the instruction set sent to the vehicle chassis electronic control unit is concise and free of logical conflicts when the transition animation ends and the system enters a fully active state. This effectively prevents the vehicle actuators from receiving oscillating control signals due to the user repeatedly switching modes during hesitation (such as rapidly switching between sand and mud), thereby protecting the vehicle chassis hardware and ensuring that the final execution state matches the user's final intention. Figure 1 To.

[0098] See attached document Figure 4 When the cockpit domain controller detects that the value of the global timer has reached or exceeded the total transition duration Ttotal, it immediately triggers the instruction batch execution subroutine. The cockpit domain controller first performs an atomic locking operation on the instruction buffer queue stored in the random access memory, pausing the reception of new pre-operation instructions to prevent data races or pointer out-of-bounds errors during the reading process.

[0099] The cockpit domain controller traverses the locked instruction buffer queue. For each instruction data packet in the queue, the cockpit domain controller invokes the protocol conversion module. The protocol conversion module is responsible for mapping the general instruction format defined by the user interface layer to the standard protocol data units required by the vehicle's underlying control network. Specifically, based on the functional domain identifier and parameter payload in the instruction data packet, the cockpit domain controller queries a pre-set signal mapping database to extract the corresponding Controller Area Network Message Identifier (CANID) and byte bit definitions. The cockpit domain controller converts the higher-level parameter values ​​(such as sand mode) into hexadecimal control words and calculates the cyclic redundancy check code according to the verification algorithm defined by the vehicle communication matrix, filling it into the data field of the message.

[0100] To avoid load congestion or packet loss on the vehicle controller area network bus caused by sudden high-density data flows, the cockpit domain controller employs an equal-slot discrete transmission strategy. Instead of sending all converted messages simultaneously, the cockpit domain controller assigns a precise transmission timestamp to each message in the queue. For each instruction in the queue, its physical transmission time on the bus is set to the product of the transition animation end time, the instruction's sequence number, and the preset minimum message transmission interval, plus a small random jitter time.

[0101] After completing the transmission of all commands in the command buffer queue, or when the cockpit domain controller detects that the command buffer queue is empty (i.e., the user has not performed any operation during the transition), the cockpit domain controller executes the default configuration loading logic. If the queue is empty, the cockpit domain controller directly reads the preset automatic off-road mode configuration parameters, encapsulates them into a control message, and sends it.

[0102] After sending a control message, the cockpit domain controller enters a closed-loop confirmation waiting state. The cockpit domain controller listens for status response messages from the vehicle chassis electronic control unit. Only when the cockpit domain controller receives a response message containing a successful execution status bit will it update the final visual state of the corresponding functional component on the display screen, switching the pre-response style of the UI interface to the activated steady-state style (e.g., changing the dashed aperture to a solid, constantly lit aperture). If no confirmation response is received within a preset timeout period, the cockpit domain controller will trigger a fault rollback mechanism, displaying a configuration failure message on the screen and reverting to the state before the switch, thus ensuring strict consistency between the information displayed on the human-machine interface and the vehicle's physical execution state.

[0103] The intelligent cockpit domain controller uses a heterogeneous multi-core system-on-a-chip (SoC) as its computing core. This SoC integrates a central processing unit, a graphics processing unit, and a neural network processing unit in its hardware architecture. To meet the demands of high-precision 3D rendering and high-reliability vehicle control during off-road mode switching, the intelligent cockpit domain controller incorporates hardware isolation technology, dividing the SoC into an application processing subsystem and a real-time safety subsystem.

[0104] The application processing subsystem is primarily responsible for running the operating system, including the 3D rendering engine. The graphics processing unit is configured to support high-concurrency floating-point operations, specifically for handling the real-time rendering of complex terrain meshes, dynamic lighting, and particle effects in off-road scenes. To prevent frame rate drops due to sudden increases in graphics load during mode switching, the intelligent cockpit domain controller incorporates a hardware-based dynamic voltage and frequency adjustment controller.

[0105] The dynamic voltage and frequency adjustment controller samples the load counter of the graphics processing unit in real time and predicts the rendering load of the next frame based on a preset off-road scene rendering complexity model. The cockpit domain controller first predicts the rendering load of the next frame based on a weighted sum of the number of vertices, the number of fragments, and the texture sampling bandwidth requirements; then, based on the base frequency, it superimposes an increment determined by the frequency adjustment gain coefficient, the difference between the predicted load and the threshold, and the chip junction temperature thermal capacity limitation coefficient to calculate the target operating frequency of the graphics processing unit.

[0106] The safety real-time subsystem comprises an independent microcontroller unit with its own static random access memory and clock source, operating independently of the application processing subsystem. The safety real-time subsystem communicates with the application processing subsystem via an internal high-speed on-chip bus, specifically responsible for executing the instruction buffer queue management and controller area network (CAN) message transmission mentioned in the previous embodiments. This hardware architecture ensures that even if the 3D rendering engine experiences lag or crashes due to extreme conditions, the chassis control command transmission logic can still be independently completed by the safety real-time subsystem, guaranteeing vehicle driving safety.

[0107] The intelligent cockpit domain controller is also equipped with a unified memory architecture. The system main memory is divided into a dedicated area for the operating system, a dedicated area for graphics memory, and a shared data exchange area. When loading scene resources, the central processing unit (CPU) uses the direct memory access controller (DRAM) to directly transfer off-road terrain data from the storage medium to the shared data exchange area without needing to copy it through the CPU's registers. The graphics processing unit (GPU) directly reads texture and model data from the shared data exchange area through virtual address mapping.

[0108] To ensure the timeliness of user operation commands in off-road mode, the intelligent cockpit domain controller integrates an Ethernet interface supporting Time-Sensitive Networking (TSN) standards and a flexible data rate interface for the controller area network (CLAN) at the physical layer. The intelligent cockpit domain controller calculates the real-time throughput bandwidth utilization of the bus interface, which is the percentage of the sum of the products of the packet size and transmission frequency of all currently active communication nodes divided by the maximum theoretical bandwidth of the physical bus. When the intelligent cockpit domain controller detects that the bandwidth utilization exceeds a safety threshold, it will activate a hardware-level Quality of Service (QoS) priority arbitration mechanism, forcibly suspending the transmission of entertainment media data streams and prioritizing the physical channel occupancy for off-road mode switching control messages.

[0109] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An immersive off-road mode switching method, characterized in that, Running within an in-vehicle human-machine interface system, the in-vehicle human-machine interface system includes a cockpit domain controller, a display screen, input devices, and a vehicle bus interface. The switching method includes the following steps. S1. The cockpit domain controller, in response to the mode switching command received through the input device, preloads the resources of the off-road driving scene in the background and controls the virtual camera to move along the planned path from the currently displayed 3D scene to the off-road driving scene in the 3D virtual environment. S2. The cockpit domain controller determines the resource readiness time threshold based on the loading progress of core resources in the off-road driving scenario and sets the total transition time. S3. When the current transfer time is between the resource ready time threshold and the total transfer time, the cockpit domain controller enters the pre-interaction state, controls the off-road core instrument components to display on the screen and opens the input interface, stores the received pre-operation instructions in the instruction buffer queue, and temporarily does not send physical control signals through the vehicle bus interface. S4. When the current transfer time reaches or exceeds the total transfer time, the cockpit domain controller enters a fully activated state, reads the pre-operation instructions in the instruction buffer queue, and sends them to the vehicle chassis electronic control unit through the vehicle bus interface.

2. The immersive off-road mode switching method according to claim 1, characterized in that, The step S1, which involves controlling the virtual camera to move along the planned path, includes: The cockpit domain controller sets a first spatial control point vector for controlling the tangential direction of the starting segment of the path and a second spatial control point vector for controlling the tangential direction of the ending segment of the path. The cockpit domain controller calculates normalized time parameters based on the ratio of the current transition time to the total transition time. The cockpit domain controller uses a cubic Bézier curve algorithm to perform weighted calculations on the initial coordinate vector of the virtual camera, the main viewpoint coordinate vector of the target off-road driving scene, the first spatial control point vector, and the second spatial control point vector according to the normalized time parameters, to obtain the instantaneous position vector of the virtual camera.

3. The immersive off-road mode switching method according to claim 1, characterized in that, Also includes: When the current transition time is less than the resource readiness time threshold, the cockpit domain controller enters a pure guidance state. In the pure guidance state, the cockpit domain controller only renders the environmental background changes caused by the movement of the virtual camera, sets the off-road core instrument components to a hidden state, and disables the interactive event response of the input device.

4. The immersive off-road mode switching method according to claim 1, characterized in that, The step S3, which involves storing the received pre-operation instructions in the instruction buffer queue, includes: The cockpit domain controller parses the pre-operation instructions into instruction data packets containing functional domain identifiers; The cockpit domain controller traverses the current instruction buffer queue to determine whether there is an old instruction data packet whose functional domain identifier is the same as the functional domain identifier of the newly generated instruction data packet. If so, the cockpit domain controller performs an overwrite operation, removing the old instruction data packet from the instruction buffer queue and inserting the newly generated instruction data packet into the instruction buffer queue; If not, the cockpit domain controller directly appends the newly generated instruction data packet to the instruction buffer queue.

5. The immersive off-road mode switching method according to claim 1, characterized in that, The S3 step of controlling the display of the off-road core instrument component on the screen includes: The cockpit domain controller calculates the progress parameters of the current transition time relative to the start of the pre-interaction state; Based on the progress parameters, the cockpit domain controller dynamically increases the transparency channel value and scaling ratio value of the off-road core instrument component using a linear interpolation algorithm, so that the off-road core instrument component gradually fades into the display as the virtual camera moves.

6. The immersive off-road mode switching method according to claim 1, characterized in that, The step S4, which involves sending data to the vehicle chassis electronic control unit via the vehicle bus interface, includes: The cockpit domain controller performs a locking operation on the instruction buffer queue and calls the protocol conversion module to convert the instruction data packets in the queue into standard protocol data units; The cockpit domain controller writes the converted standard protocol data units sequentially into the network transmission buffer according to the timestamp order generated by the instructions. If the instruction buffer queue is detected to be empty, the cockpit domain controller generates a preset default configuration instruction and sends it to the vehicle chassis electronic control unit.

7. The immersive off-road mode switching method according to claim 1, characterized in that, The pre-interaction state also includes: When the cockpit domain controller detects a touch event targeting the off-road core instrument component, it initiates the visual response thread without waiting for confirmation from the vehicle chassis electronic control unit. The visual response thread modifies the rendering material properties of the touched target control object, adjusts its self-illumination intensity, and performs local geometric deformation on the target control object to provide visual feedback on the user's pre-operation.

8. The immersive off-road mode switching method according to claim 2, characterized in that, Step S1, which involves controlling the virtual camera to move along the planned path in the 3D virtual environment, also includes: The cockpit domain controller acquires the vehicle's real-time driving speed and calculates dynamic field of view parameters based on the real-time driving speed, so that the field of view increases with the increase of vehicle speed; The cockpit domain controller acquires the vehicle's vertical acceleration signal, generates a pseudo-random three-dimensional vector using a noise generation function, and superimposes the pseudo-random three-dimensional vector as a vibration offset into the instantaneous position vector of the virtual camera.

9. The immersive off-road mode switching method according to claim 1, characterized in that, The steps of preloading resources for off-road driving scenarios in the background include: The cockpit domain controller adopts an incremental asynchronous loading strategy, performing data decompression and video memory upload within the allowable range of rendering time for each frame; The cockpit domain controller calculates the loading position of the off-road driving scene in the world coordinate system based on the current camera position and line of sight. The loading position is the coordinate after extending a preset distance along the line of sight and superimposing a vertical offset vector.

10. The immersive off-road mode switching method according to claim 6, characterized in that, The step S4, which involves sending data to the vehicle chassis electronic control unit via the vehicle bus interface, also includes: The cockpit domain controller monitors the real-time throughput and bandwidth utilization of the vehicle bus interface; When the real-time throughput bandwidth utilization exceeds the preset safety threshold, the cockpit domain controller triggers the quality of service priority arbitration mechanism, suspends the transmission of non-critical media data streams, and prioritizes the transmission of standard protocol data units in the instruction buffer queue.

Images