Vehicle running control method and control system based on driving assistance and vehicle

Abstract

The invention discloses a vehicle running control method and system based on driving assistance and a vehicle thereof, and belongs to the technical field of vehicle driving assistance. The vehicle running control method is achieved based on vehicle driving assistance, and when a driving assistance mode of the vehicle is started, road condition information of the current vehicle is obtained; the method comprises the following steps: acquiring road condition information of a vehicle, performing scene semantic analysis by utilizing the road condition information, extracting key parameters corresponding to the road condition information, performing global optimization decision according to the key parameters and the current driving state of the vehicle, calculating to obtain an optimal control sequence of the vehicle, and determining an optimal control mode according to the lowest energy consumption; and finally, a power system and a braking system for controlling the vehicle are output according to the optimal control sequence, so that the vehicle executes speed regulation operation to reach a target state, global cooperative work can be achieved, the control mode of the whole vehicle is optimized, the purpose of reducing energy consumption is achieved, and the cruising ability of the vehicle is improved.

Description

Technical Field

[0001] This invention relates to the field of vehicle driver assistance technology, and in particular to a vehicle driving control method, control system, vehicle controller and vehicle based on driver assistance. Background Technology

[0002] In recent years, the development of automotive driver assistance has been rapid. Among the related technologies, signals provided by sensors are commonly used, such as the distance to the vehicle in front and speed limit signs. These signals are used to control the vehicle to perform acceleration or deceleration operations. However, the control is not well integrated with the vehicle's energy, resulting in significant energy loss and hindering the improvement of the vehicle's range. Summary of the Invention

[0003] The present invention aims to solve the technical problems existing in the above-mentioned related technologies, and proposes a vehicle driving control method based on driving assistance, which can optimize the whole vehicle control in a coordinated manner across the entire domain, effectively reduce energy loss, and improve the vehicle's range.

[0004] The present invention also provides a control system, a vehicle controller and a vehicle thereof that apply the above-described vehicle driving control method based on driving assistance.

[0005] According to a first aspect of the present invention, a vehicle driving control method based on driving assistance includes: When the vehicle's driver assistance mode is activated, it obtains the current road condition information. The road condition information is used to perform scene semantic parsing in order to extract the key parameters corresponding to the road condition information; Based on the key parameters and the current driving state of the vehicle, a global optimization decision is made to calculate the optimal control sequence of the vehicle. The optimal control sequence is used to characterize the control method corresponding to the lowest energy consumption when the vehicle changes from the current driving state to the target state. The optimal control sequence is used to control the vehicle's power system and braking system, enabling the vehicle to perform speed adjustment operations to achieve the target state.

[0006] The vehicle driving control method based on driving assistance according to embodiments of the present invention has at least the following beneficial effects: The vehicle driving control method in this embodiment is based on automotive driving assistance. When the vehicle is in driving assistance mode, it acquires the current road condition information and uses this information for scene semantic parsing to extract key parameters corresponding to the road condition information. This converts complex road condition information into semantic information that is easily identifiable. Then, based on the key parameters and the vehicle's current driving state, it performs a global optimization decision to calculate the optimal control sequence for the vehicle. This sequence predicts the minimum energy consumption required for the vehicle to change from its current driving state to the target state, and determines the optimal control method based on this minimum energy consumption. Finally, it outputs the optimal control sequence to control the vehicle's power system and braking system, enabling the vehicle to perform speed adjustment to reach the target state. Compared to the control methods in related technologies, the vehicle driving control method in this embodiment does not rely solely on collecting road condition information to control vehicle acceleration or deceleration. Instead, it integrates vehicle energy consumption with control, simultaneously controlling the vehicle's power system and braking system to execute a unified optimal control sequence. This achieves global collaborative operation, optimizes the overall vehicle control method, reduces energy consumption, and thus improves the vehicle's range.

[0007] According to some embodiments of the present invention, the road condition information includes information about the vehicle ahead, traffic light information, and road sign information; The step of using the road condition information to perform scene semantic parsing to extract key parameters corresponding to the road condition information includes: The information of the vehicle in front, the traffic light information, and the road sign information are fused from multiple sources to perform scene recognition and classification, and the key parameters are extracted. The key parameters include the distance to the vehicle in front, the traffic light time, and the slope of the road surface.

[0008] According to some embodiments of the present invention, the step of performing global optimization decision-making based on the key parameters and the current driving state of the vehicle to calculate the optimal control sequence of the vehicle includes: The minimum energy consumption is calculated based on the key parameters and the current driving state using a real-time optimization algorithm based on dynamic programming or model predictive control. The optimal control sequence is obtained based on the minimum energy consumption and the overall energy consumption of the vehicle.

[0009] According to some embodiments of the present invention, the current driving state includes the vehicle speed, the vehicle's battery state of charge, and the battery's health status. The real-time optimization algorithm based on dynamic programming or model predictive control calculates the minimum energy consumption according to the key parameters and the current driving state, including: Retrieve the pre-constructed cost function; Using the distance to the vehicle in front, the timing of the traffic light, the slope of the road surface, the vehicle speed, the vehicle's battery state of charge, and the battery's health status as parameters, the minimum energy consumption is calculated based on these parameters and the cost function.

[0010] According to some embodiments of the present invention, the step of performing global optimization decision-making based on the key parameters and the current driving state of the vehicle to calculate the optimal control sequence of the vehicle includes: When the traffic light is red, the optimal control sequence for the vehicle is determined according to the real-time optimization algorithm. The optimal control sequence includes the vehicle's optimal deceleration start point and curve, as well as a set time. The step of outputting control of the vehicle's power system and braking system according to the optimal control sequence includes: Based on the optimal deceleration start point and curve, the power system is controlled to output zero torque and enter a coasting state; After the vehicle has been in the coasting state for the set time, the braking system is controlled to brake the vehicle. According to a second aspect of the present invention, a vehicle driving control system based on driving assistance includes: The data acquisition module is used to obtain the current road condition information of the vehicle when the driving assistance mode is activated; A scene semantic parser is used to perform scene semantic parsing using the road condition information in order to extract the key parameters corresponding to the road condition information; The global optimization controller is used to make global optimization decisions based on the key parameters and the current driving state of the vehicle, and calculate the optimal control sequence. The optimal control sequence is used to characterize the control mode corresponding to the lowest energy consumption for the vehicle to change from the current driving state to the target state. The controller outputs the control of the vehicle's power system and braking system according to the optimal control sequence, so that the vehicle performs speed adjustment operation to achieve the target state.

[0011] The vehicle driving control system based on driving assistance according to embodiments of the present invention has at least the following beneficial effects: The vehicle driving control system in this embodiment is based on automotive driving assistance. It utilizes a data acquisition module to acquire current road condition information when the vehicle's driving assistance mode is activated. A scene semantic parser then uses this road condition information to perform scene semantic analysis, extracting key parameters corresponding to the road condition information. This transforms complex road condition information into easily identifiable semantic information. A global optimization controller then performs global optimization decisions based on these key parameters and the vehicle's current driving state, calculating the optimal control sequence for the vehicle. This sequence predicts the minimum energy consumption required for the vehicle to transition from its current driving state to the target state, and determines the optimal control method based on this minimum energy consumption. Finally, the optimal control sequence is used to control the vehicle's power system and braking system, enabling the vehicle to perform speed adjustments to reach the target state. Compared to the control methods in related automotive driving assistance technologies, the vehicle driving control system in this embodiment does not solely rely on acquiring road condition information to control vehicle acceleration or deceleration. Instead, it integrates vehicle energy consumption with control, simultaneously controlling the vehicle's power system and braking system to execute a unified optimal control sequence. This achieves global collaborative operation, optimizes the overall vehicle control method, reduces energy consumption, and thus improves the vehicle's range.

[0012] According to some embodiments of the present invention, the road condition information includes information about the vehicle ahead, traffic light information, and road sign information; The scene semantic parser is also used to perform scene recognition and classification by fusing the information of the vehicle in front, the traffic light information, and the road sign information through multi-source data, and to extract the key parameters, which include the distance to the vehicle in front, the traffic light time, and the slope of the road surface. The global optimization controller is also used to calculate the minimum energy consumption based on the key parameters and the current driving state using a real-time optimization algorithm based on dynamic programming or model predictive control; and to obtain the optimal control sequence based on the minimum energy consumption and the overall energy consumption of the vehicle.

[0013] According to a third aspect of the present invention, a vehicle controller includes at least one processor; and a memory storing instructions that, when executed by the at least one processor, perform the vehicle driving control method based on driving assistance described in the first aspect of the present invention.

[0014] According to a fourth aspect of the present invention, a vehicle includes the vehicle driving control system based on driving assistance described in the second aspect of the above embodiments, or the vehicle controller described in the third aspect of the above embodiments.

[0015] Since the vehicle adopts all the technical solutions of the vehicle driving control system or vehicle controller based on driving assistance in the above embodiments, it has at least all the beneficial effects brought about by the technical solutions in the above embodiments, which will not be repeated here.

[0016] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. Attached Figure Description

[0017] Figure 1 This is a flowchart of a vehicle driving control method based on driving assistance according to an embodiment of the present invention; Figure 2 This is a flowchart illustrating a specific example of obtaining the optimal control sequence for a vehicle in one embodiment of the present invention. Figure 3 This is a flowchart illustrating a specific example of calculating the minimum energy consumption in one embodiment of the present invention; Figure 4 This is a flowchart illustrating a specific example of a vehicle encountering a traffic light in one embodiment of the present invention. Figure 5 This is a flowchart illustrating a specific example of a vehicle driving control method based on driving assistance according to an embodiment of the present invention. Detailed Implementation

[0018] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0019] In the description of this invention, "multiple" means two or more; "greater than," "less than," and "exceeding" are understood to exclude the stated number; "above," "below," and "within" are understood to include the stated number. The use of "first" and "second" in the description is for the purpose of distinguishing technical features only and should not be construed as indicating or implying relative importance, or implicitly indicating the number of indicated technical features, or implicitly indicating the order of the indicated technical features.

[0020] In the description of this invention, it should be noted that terms such as "set up," "install," and "connect" should be interpreted broadly, and those skilled in the art can reasonably determine the specific meaning of the above terms in this invention in conjunction with the specific content of the technical solution.

[0021] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are some embodiments of the present invention, not all embodiments.

[0022] In related technologies, automotive driver assistance systems primarily rely on sensor signals for single-dimensional vehicle control. Traditional methods directly trigger acceleration or deceleration by recognizing the distance to the vehicle ahead or a speed limit sign, without systematically correlating the vehicle's operating state with the surrounding environment and energy consumption. This isolated control mode results in a lack of a global perspective in energy management. For example, when approaching traffic lights, the vehicle may frequently perform unnecessary braking operations, leading to low kinetic energy recovery efficiency and impacting overall range performance.

[0023] To address the aforementioned issues, this application proposes a vehicle driving control method based on driver assistance, which can coordinately optimize vehicle control across the entire domain. When the vehicle is in driver assistance mode, road condition information is acquired, key parameters are extracted through scene semantic parsing, and the optimal control sequence is generated by making a global optimization decision based on the current state of the vehicle. Finally, the power system and braking system are coordinated to perform speed adjustment operations. This technical solution can effectively reduce energy loss and improve the vehicle's range.

[0024] Reference Figure 1 As shown, the vehicle driving control method based on driving assistance in this embodiment includes the following steps: Step S100: When the vehicle activates the driving assistance mode, obtain the current road condition information of the vehicle. Step S200: Use road condition information to perform scene semantic parsing in order to extract the key parameters corresponding to the road condition information; Step S300: Perform global optimization decision based on key parameters and the current driving state of the vehicle, and calculate the optimal control sequence of the vehicle. The optimal control sequence is used to characterize the control mode corresponding to the lowest energy consumption when the vehicle changes from the current driving state to the target state. Step S400: Output the power system and braking system of the vehicle according to the optimal control sequence to make the vehicle perform speed adjustment operation to achieve the target state.

[0025] Understandably, road condition information can be collected in real time by vehicle sensors and vehicle-to-everything (V2X) systems to collect data on the vehicle's surrounding environment. For example, millimeter-wave radar can be used to detect the distance to the vehicle in front, cameras can be used to identify the status of traffic lights and road signs, and GPS positioning can be used to obtain road condition information, providing multi-dimensional environmental parameters for subsequent optimization.

[0026] Scene semantic parsing can be understood as transforming sensor-collected data into quantifiable control parameters. Its purpose is to convert complex environmental features into easily identifiable semantic information about road conditions, achieving scene perception and semantic parsing. Specifically, multi-source data fusion algorithms can be used to comprehensively analyze information such as the trajectory of the vehicle ahead, traffic light countdowns, and road inclination angles to establish a semantic model of the current driving scenario. For example, the input driving assistance raw data includes real-time information such as the vehicle ahead, traffic lights, and road signs, and the output is structured scene semantic labels and their key parameters. Key parameters can be distance parameters to the vehicle ahead, traffic light timing parameters, and road markings, etc.

[0027] Understandably, global optimization decision-making can employ model predictive control algorithms to analyze and process key parameters and the vehicle's current driving state. The aim is to make decisions based on key parameters obtained through scene perception and semantic parsing, combined with the vehicle's current driving state, to generate the optimal control sequence. For example, it can continuously calculate the optimal acceleration curve for multiple future time steps within the time domain to ensure minimal energy consumption throughout the entire process. The optimal control sequence can be a transformation of theoretical optimization results into system commands for execution control. For instance, a torque distribution controller can coordinate the output of the drive motor and the timing of mechanical braking intervention to achieve seamless integration between the power system and the braking system.

[0028] To illustrate with a specific example, when the vehicle enters driver assistance mode, the environmental perception system continuously collects road information within a 300-meter radius ahead. Through multi-sensor data fusion, it identifies the traffic situation at the upcoming intersection with 20 seconds remaining on the red light. At this time, the vehicle is traveling at 60 km / h, and the battery's state of charge is 50%. Based on the current vehicle speed, battery state, and the target stop line position, the model predictive control algorithm calculates the optimal control sequence: starting coasting 150 meters in advance and applying braking in the final 20 meters. The powertrain gradually reduces drive torque according to this optimal control sequence, utilizing vehicle inertia to reduce energy consumption, while the braking system intervenes at the predetermined position to achieve a precise stop. This entire process reduces energy loss compared to traditional emergency braking methods and avoids efficiency losses caused by frequent battery charging and discharging.

[0029] Traditional control methods typically calculate the safe braking distance after recognizing a red light and then control the vehicle to brake. However, this invention, unlike traditional control methods, does not rely solely on collecting road condition information to control vehicle acceleration or deceleration. Instead, it integrates vehicle energy consumption with control. For example, it considers the impact of road gradient on coasting distance, the current charging and discharging efficiency characteristics of the battery, and changes in traffic light timing. Through multi-parameter coupling optimization, it selects the deceleration strategy with the highest energy recovery efficiency. At the same time, it achieves coordinated operation of the power system and braking system through a unified control sequence. For example, on long downhill sections, it prioritizes using gradient potential energy to maintain vehicle speed, reducing driving energy consumption and achieving full-domain coordinated operation. This optimizes the overall vehicle control method, reduces energy consumption, and thus improves the vehicle's range.

[0030] It should be noted that this application enables global energy optimization control of vehicles in complex road scenarios. By establishing the correlation between the environmental semantic model and vehicle dynamics, it effectively coordinates the working modes of the power system and braking system, significantly reducing unnecessary energy loss while ensuring driving safety. Especially in urban road conditions with frequent start-stop cycles, it can intelligently select the deceleration strategy with the highest kinetic energy recovery efficiency, increasing the electric vehicle's range by approximately 5%-8% while reducing wear on mechanical braking components.

[0031] Reference Figure 2 As shown, in some embodiments, steps S200 and S300 specifically include the following steps: Step S210: The information of the vehicle in front, traffic light information, and road sign information are fused from multiple sources to perform scene recognition and classification, and key parameters are extracted. Step S310: Based on the real-time optimization algorithm of dynamic programming or model predictive control, the minimum energy consumption is calculated according to the key parameters and the current driving state. Step S320: Obtain the optimal control sequence based on the lowest energy consumption and the overall energy consumption of the vehicle.

[0032] Understandably, road condition information includes information about vehicles ahead, traffic lights, and road signs. Information about vehicles ahead can be obtained through onboard sensors that capture dynamic data of obstacles in front of the vehicle, specifically using millimeter-wave radar or visual cameras, to monitor the position and relative speed of vehicles ahead in real time. Traffic light information refers to the current state and remaining time of traffic lights, which can be obtained through vehicle-to-infrastructure communication modules or image recognition algorithms to determine the cycle constraints of traffic lights. Road sign information refers to static signage information set up in the road environment, which can be obtained through high-precision map matching or visual recognition technology to analyze features such as speed limits and gradients. Multi-source data fusion involves spatiotemporal alignment and correlation analysis of heterogeneous data acquired from different sensors, specifically using Kalman filtering or deep learning models to eliminate errors from single data sources and improve scene recognition accuracy.

[0033] Specifically, after the vehicle activates driver assistance mode, the system simultaneously collects three types of data: distance to the vehicle ahead, traffic light countdown timer, and road conditions. The distance to the vehicle ahead is continuously updated via a radar ranging module; the traffic light countdown timer receives data from the intersection signal controller via a wireless communication module; and road conditions are acquired through high-precision map matching or inertial sensors. This data, after timestamp alignment, is input into a fusion algorithm, such as a Bayesian network-based probabilistic model, to calculate the confidence level for different scenario categories. When a following scenario is identified, the system prioritizes extracting the distance to the vehicle ahead; when a waiting scenario at an intersection is identified, it prioritizes extracting the remaining traffic light time; and when a driving scenario on a slope is identified, it extracts the road slope parameter. Through a dynamic weighted fusion strategy, the system automatically adjusts the weight coefficients of each parameter to ensure that the extraction accuracy of key parameters meets the subsequent energy consumption optimization requirements.

[0034] Traditional methods typically rely on a single type of road condition information for control decisions, such as adjusting vehicle speed solely based on the distance to the vehicle ahead, without considering the impact of traffic light cycles on coasting distance, or ignoring the effect of road gradient on energy recovery efficiency. This invention, however, integrates multi-dimensional road condition information, establishing a dynamic correlation model of traffic participants, traffic signal constraints, and road terrain features during the scene analysis phase. This allows subsequent optimization algorithms to calculate control strategies based on more comprehensive environmental parameters, avoiding frequent braking or ineffective acceleration due to missing scene features.

[0035] This application enables the accurate identification of key feature parameters of the current scene during vehicle operation, providing precise input conditions for energy optimization control. For example, when approaching a red light intersection, the system can calculate the optimal coasting distance by combining the remaining traffic light time with the current vehicle speed, avoiding energy waste caused by premature braking. On inclines and declines, the system adjusts the power output curve based on the gradient parameters to reduce ineffective motor power consumption. This scene analysis method based on multi-source data fusion effectively improves the accuracy of energy consumption optimization calculations, thereby reducing overall vehicle energy loss.

[0036] Dynamic Programming (DP) is a mathematical method that recursively solves for the optimal solution in a multi-stage decision-making process. Specifically, it can be implemented using state transition equations and iterative value function calculations, and is suitable for scenarios where multi-stage energy consumption optimization is required during vehicle operation. Model Predictive Control (MPC) is a closed-loop control method based on rolling time-domain optimization. Specifically, it can be implemented using predictive models and online optimization calculations, and can dynamically adjust the control strategy based on real-time updated road condition information. Real-time optimization algorithms are computational processes aimed at reducing energy consumption and solving for the optimal control sequence within a finite time. Specifically, they can be implemented using convex optimization or heuristic search algorithms to ensure that the vehicle quickly generates feasible solutions in complex scenarios. Comprehensive energy consumption data includes a multi-dimensional parameter set encompassing battery state of charge, battery health status, and historical energy consumption data. Specifically, it can be collected and fused by the onboard battery management system to balance the correlation between instantaneous energy consumption and long-term battery performance.

[0037] Specifically, during vehicle operation, the real-time optimization algorithm, based on dynamic programming or model predictive control frameworks, inputs key parameters such as distance to the vehicle ahead, traffic light duration, road gradient, and current driving conditions including vehicle speed, battery state of charge, and battery health into a pre-constructed cost function. The cost function aims to minimize energy consumption and, combining vehicle dynamics and energy conversion efficiency models, iteratively calculates and generates an optimal control sequence that meets safety constraints and comfort requirements. For example, when the vehicle approaches a red light, model predictive control predicts speed changes over the next few seconds through rolling optimization, dynamically adjusting the coordination strategy of coasting and braking to maximize kinetic energy recovery before the vehicle stops. Simultaneously, for long-distance driving scenarios, the dynamic programming algorithm divides the entire journey into multiple stages and determines the optimal acceleration and energy allocation for each stage through global optimization, thereby reducing overall energy consumption.

[0038] Reference Figure 3 As shown, in some embodiments, step S310 specifically includes the following steps: Step S311: Retrieve the pre-constructed cost function; Step S312: Using the distance to the vehicle in front, the traffic light time, the road slope, the vehicle speed, the vehicle's battery state of charge, and the battery's health status as parameters, calculate the minimum energy consumption based on the parameters and the cost function.

[0039] The cost function is a mathematical expression used to quantify system performance. Specifically, it can integrate multiple objective parameters such as energy consumption, driving time, and comfort using a weighted summation method, with parameter adjustments reflecting the priority of different optimization objectives. The State of Charge (SOC) is the ratio of the battery's remaining usable capacity to its rated capacity. It can be estimated in real-time using the ampere-hour integral method or the Kalman filter algorithm, and is used to characterize the vehicle's current energy reserve level. The State of Health (SOH) is the ratio of the battery's current maximum usable capacity to its initial capacity. It can be estimated using an aging model built from cyclic charge-discharge test data, and is used to reflect the degree of battery performance degradation.

[0040] Specifically, during vehicle operation, onboard sensors continuously collect data on the distance to the vehicle ahead, the remaining time of traffic lights, and road slope parameters, while simultaneously monitoring vehicle speed, battery state of charge, and battery health. A pre-built cost function is retrieved, and these parameters are input into dynamic programming or model predictive control algorithms for comprehensive calculation. For example, when a red light is detected approaching, the algorithm combines the current vehicle speed, remaining battery charge, and battery aging level to calculate a deceleration curve that maximizes braking energy recovery while minimizing battery wear, while maintaining a safe distance. Battery health parameters serve as constraints to prevent deep discharge from accelerating battery aging. Through multi-parameter coupling optimization, optimal control commands are generated that balance immediate energy consumption and long-term battery life.

[0041] Traditional methods typically make control decisions based solely on instantaneous road condition parameters, failing to incorporate battery health status into the optimization model. This leads to accelerated battery performance degradation over long-term use, and their energy consumption calculations are often limited to current driving conditions, lacking a comprehensive consideration of the battery's entire lifecycle. In contrast, this invention introduces battery health status parameters and establishes a correlation model between battery aging and energy consumption during the optimization process. This allows the control strategy to not only reduce current energy consumption but also effectively slow down battery capacity degradation.

[0042] Through the above technical solution, this application achieves the integration of vehicle control strategy and battery life management. While ensuring driving safety, it generates an optimal control sequence that balances immediate energy efficiency and long-term battery health by accurately calculating the impact of multi-dimensional parameters on energy consumption. This not only reduces energy loss during a single trip but also extends battery life by optimizing battery usage patterns, thereby achieving continuous energy consumption reduction throughout the vehicle's lifespan.

[0043] Reference Figure 4 As shown, in some embodiments, steps S300 and S400 specifically include the following steps: Step S330: When the traffic light is red, determine the optimal control sequence of the vehicle according to the real-time optimization algorithm. The optimal control sequence includes the optimal deceleration start point and curve of the vehicle and the set time. Step S410: Based on the optimal deceleration start point and curve, control the power system to output zero torque and enter the coasting state; In step S420, after a set time has elapsed while the vehicle is coasting, the braking system is controlled to brake the vehicle.

[0044] The optimal control sequence can be understood as a set of control commands generated by dynamic programming or model predictive control algorithms, including the deceleration start point, speed change curve, and time parameters. Specifically, it can be implemented using a rolling time-domain optimization method to coordinate the timing of the powertrain and braking systems in red-light scenarios. The real-time optimization algorithm is a decision-making method based on online calculations of the current vehicle state and road conditions. Specifically, it can be implemented using a constrained quadratic programming solver to dynamically adjust the deceleration start point and coasting time. Coasting state refers to the condition where the powertrain stops outputting torque and the vehicle moves due to inertia. This can be achieved by disconnecting the powertrain drivetrain or adjusting the motor torque to zero, thereby reducing powertrain energy consumption.

[0045] Specifically, when a red traffic light is detected ahead, a real-time optimization algorithm calculates the distance between the vehicle's current position and the red light stop line, its current speed, and road gradient parameters to determine the optimal deceleration starting point. Upon reaching the deceleration starting point, the powertrain immediately cuts off torque output, allowing the vehicle to enter a coasting state. During coasting, the vehicle speed naturally decreases due to road gradient and air resistance until a preset time threshold is reached, triggering the braking system. The braking system applies progressive braking force based on the relationship between the remaining coasting distance and the target stopping position to complete the stopping operation. The entire process maximizes the use of the vehicle's inertial kinetic energy during the coasting phase, reducing energy loss from the powertrain's active deceleration, while time threshold control ensures that the timing of braking intervention matches the stopping accuracy.

[0046] In some embodiments, the vehicle also includes a chassis system with adjustable suspension height, which is a device that can change the vertical distance between the vehicle body and the ground via an electric or pneumatic actuator. Specifically, this can be achieved by using air springs in conjunction with a height sensor, which reduces the frontal area of ​​the vehicle when it is in motion by lowering the vehicle body height.

[0047] Long-distance coasting or downhill driving refers to the condition where the vehicle continues to travel without power output or driven by gravitational potential energy. This can be achieved through joint detection using inertial sensors and slope sensors, and its purpose is to identify scenarios where aerodynamic characteristics need to be optimized. The chassis system shortens the suspension travel to a preset height range, which can be achieved by using solenoid valves to control the exhaust of air springs, thereby reducing air resistance during vehicle movement.

[0048] Adjusting the suspension damping refers to changing the flow resistance of the hydraulic oil inside the shock absorber. Specifically, this can be achieved by adjusting the damping coefficient through an electronically controlled damping adjustment valve. Its function is to suppress the mechanical energy loss caused by vehicle vibration.

[0049] Specifically, when a vehicle is coasting for long distances or descending a slope, air resistance becomes the primary source of energy consumption. In this situation, lowering the vehicle's height reduces the distance between the chassis and the ground, thereby reducing the intensity of air turbulence. Simultaneously, the suspension damping coefficient is adjusted to a medium-firm mode to reduce vertical vibrations caused by road bumps and avoid energy loss from repeated compression and release of the suspension system. The combined effect of vehicle height and damping adjustments, while maintaining driving stability, optimizes aerodynamic performance and mechanical energy conversion efficiency to reduce energy consumption.

[0050] Traditional driver assistance systems only regulate vehicle speed through the braking system, neglecting the impact of the suspension system on energy utilization. This invention, by actively controlling suspension height and damping parameters, incorporates chassis dynamics into the overall vehicle energy management framework, achieving dual optimization of air resistance and mechanical losses in specific driving scenarios. Through this technical solution, this application effectively reduces air resistance during coasting or downhill driving, while simultaneously minimizing energy loss due to suspension system vibration, thereby improving overall vehicle energy efficiency and extending vehicle range.

[0051] Reference Figure 5 As shown below, a specific example illustrates the vehicle driving control method. After sensors collect information such as vehicles, traffic lights, and road signs, scene perception and semantic parsing are performed. Scene recognition and classification are conducted through multi-source data fusion, and key parameters such as distance, time, and slope are extracted. Combined with the vehicle's current state (speed, SOC, SOH), a global optimization decision is made. First, a cost function is constructed to calculate and minimize the total energy consumption. The optimal control sequence is then solved based on the energy consumption situation. The control sequence is then released to the power system (controlling motor torque, drive recovery state, etc.), the braking system (controlling electric braking and hydraulic braking), and the chassis system (suspension height, damping, and air kit state) to enable the vehicle to smoothly perform deceleration and acceleration.

[0052] The specific implementation process of the vehicle driving control method includes three stages: stage one, stage two, and stage three. Stage one involves scene perception and semantic parsing; stage two involves global optimization decision-making to obtain the optimal control sequence; and stage three involves coordinated execution and feedback to enable the vehicle to accelerate or decelerate.

[0053] This invention, for the first time, applies dynamic programming / MPC optimization algorithms to the real-time calculation of energy recovery strategies, elevating the process from "rule-based control" to "optimal control." It also proposes a scenario semantic parsing module to transform driver assistance information into a high-level language directly usable by the optimization algorithm. Furthermore, it incorporates the chassis system as a controllable actuator into the energy management closed loop, achieving cross-domain, truly global energy consumption collaborative optimization. Through global optimization, it can theoretically approach the minimum energy consumption under the physical limits of the road segment. The deceleration curve generated by the optimization algorithm is naturally smooth and optimal. The system behavior is no longer a simple response but rather based on an understanding and planning of future scenarios, more closely resembling the decision-making patterns of human drivers.

[0054] This invention also proposes a vehicle driving control system based on driver assistance, including a data acquisition module, a scene semantic parser, and a global optimization controller. The data acquisition module is used to acquire the current road condition information of the vehicle when the driver assistance mode is activated; the scene semantic parser is used to perform semantic parsing on the road condition information to extract key parameters; the global optimization controller is used to combine the key parameters with the current driving state of the vehicle to make global optimization decisions, generate the optimal control sequence, and output control commands to the power system and braking system to realize vehicle speed adjustment.

[0055] The data acquisition module refers to the device that acquires external environmental information through onboard sensors, specifically using one or more combinations of cameras, millimeter-wave radar, and lidar. Its function is to provide real-time traffic data support for subsequent scene analysis and optimization decisions. The scene semantic parser is the computational unit that fuses and processes multi-source heterogeneous data, specifically using deep learning-based semantic segmentation algorithms or rule engines. Its function is to transform discrete information such as distance to the vehicle ahead and traffic light status into structured key parameters. The global optimization controller is the decision-making unit that performs real-time optimization calculations, specifically using dynamic programming algorithms or model predictive control algorithms. Its function is to generate an energy-optimized control sequence based on vehicle status and key parameters. The powertrain and braking system refer to the actuators that perform speed regulation operations, specifically using motor controllers and electronic braking units. Their function is to adjust torque output and braking force distribution according to the control sequence.

[0056] Specifically, the data acquisition module can execute step S100 in the vehicle driving control method of the above embodiment, the scene semantic parser can execute steps S200 and S210 in the vehicle driving control method of the above embodiment, and the global optimization controller can execute steps S300, S400, S310, S320, etc. in the vehicle driving control method of the above embodiment.

[0057] Specifically, the data acquisition module continuously collects information on the position of the vehicle ahead, the status of traffic lights, and road slope, and transmits the raw data to the scene semantic parser. The scene semantic parser uses multi-source data fusion technology to quantify the remaining time of the traffic lights and the relative distance to the vehicle ahead, combining this with the road slope parameters to form a structured input. After receiving the structured parameters, the global optimization controller combines the current vehicle speed, battery state of charge, and health status, and uses a dynamic programming algorithm to search for the lowest energy consumption path in both the time and spatial domains, generating an optimal control sequence that includes acceleration curves and braking timing. The powertrain adjusts the motor output torque according to the control sequence, and the braking system intervenes to decelerate at predetermined times; both work together to achieve a smooth transition of the vehicle from the current state to the target state.

[0058] Traditional solutions rely solely on discrete control based on data from a single sensor, such as triggering emergency braking based only on the distance to the vehicle ahead, without considering the impact of battery status and road gradient on energy consumption. This invention constructs a global optimization model through multi-source data fusion, simultaneously integrating the powertrain efficiency characteristics and braking energy recovery potential in control decisions, avoiding energy waste caused by localized optimization. For example, in long downhill scenarios, existing technologies may continuously apply mechanical braking, leading to energy dissipation, while this invention dynamically adjusts the coasting strategy based on gradient parameters and battery status to maximize potential energy recovery.

[0059] Through the above technical solutions, this application enables coordinated control of the power system and braking system in complex traffic scenarios, dynamically adjusting the control strategy based on real-time road conditions and vehicle status. For example, in traffic light scenarios, by pre-calculating the optimal deceleration curve, the vehicle can coast to the stop line with minimal energy consumption, avoiding energy loss caused by frequent start-stop operations. In following vehicle scenarios, the acceleration or coasting strategy is selected based on the distance to the vehicle in front and the battery health status, balancing safe distance maintenance with energy efficiency. This effectively reduces the overall vehicle operating energy consumption and extends the driving range of electric vehicles.

[0060] This application further proposes that road condition information includes information about the vehicle in front, traffic light information, and road sign information; a scene semantic parser uses multi-source data fusion to perform scene recognition and classification of information about the vehicle in front, traffic light information, and road sign information, and extracts key parameters, including the distance to the vehicle in front, the timing of the traffic light, and the slope of the road surface; a global optimization controller is based on a real-time optimization algorithm of dynamic programming or model predictive control, calculates the minimum energy consumption based on the key parameters and the current driving state, and obtains the optimal control sequence based on the minimum energy consumption and the overall energy consumption of the vehicle.

[0061] Multi-source data fusion refers to integrating and cross-validating data from different sensors, including information on vehicles ahead, traffic lights, and road signs. This can be achieved using Kalman filtering or Bayesian network algorithms to eliminate errors from single data sources and improve scene recognition accuracy. Dynamic programming involves decomposing the vehicle's driving process into multiple stages and finding the global optimum. This can be achieved using Bellman equations for recursive calculations and is suitable for energy consumption optimization problems in discrete states. Model predictive control involves establishing a predictive model in the rolling time domain and solving for the optimal control quantity. This can be achieved using quadratic programming algorithms for online optimization, enabling real-time adjustments to the control strategy to adapt to continuously changing driving environments.

[0062] Specifically, the scene semantic parser uses multi-source data fusion to jointly analyze the distance to the vehicle ahead, the remaining time of the traffic light, and the road slope, forming a driving scene model with temporal and spatial constraints. For example, when the remaining time of the red light ahead is detected to be a fixed value, the parser will generate constraints including a time window. The global optimization controller calculates the deceleration curve that satisfies the time constraints and has the lowest energy consumption, based on the current vehicle speed, battery state of charge, and slope parameters, within the framework of dynamic programming or model predictive control. In long downhill scenarios, the optimization algorithm prioritizes using gravitational potential energy to maintain vehicle speed, reducing power system output; in traffic light scenarios, by planning the coasting distance and braking timing in advance, it avoids wasting kinetic energy due to sudden braking.

[0063] Furthermore, embodiments of the present invention also provide a vehicle controller, including: at least one processor; and a memory storing instructions, which, when executed by the at least one processor, execute the vehicle driving control method based on driving assistance described above.

[0064] Taking the example of a processor and memory in a vehicle controller being connected via a bus, the memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, the memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory may optionally include memory remotely located relative to the control processor, and these remote memories can be connected to the controller via a network.

[0065] The non-transient software program and instructions required to implement the vehicle driving control method of the above embodiments are stored in memory. When executed by the processor, the control method described above is executed, for example, the method described above. Figure 1 Method steps S100 to S400 Figure 2 Method steps S210 to S320, Figure 3 Method steps S311 to S312, Figure 4 The method steps S330 to S420, etc.

[0066] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0067] This invention also provides a vehicle, including the vehicle driving control method or vehicle controller based on driving assistance described above. The vehicle can be a private car, such as a sedan, SUV, MPV, or pickup truck. The vehicle can also be a commercial vehicle, such as a van, bus, small truck, or large trailer. When the vehicle is a new energy vehicle, it can be a hybrid vehicle or a pure electric vehicle.

[0068] Since the vehicle adopts all the technical solutions of the vehicle controller in the above embodiments, it has at least all the beneficial effects brought about by the technical solutions in the above embodiments, which will not be repeated here.

[0069] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

Claims

1. A vehicle travel control method based on driving assistance, characterized by, The method comprises the following steps: When the vehicle starts the driving assistance mode, the road condition information of the current vehicle is acquired; The scene semantic analysis is performed by using the road condition information to extract the key parameters corresponding to the road condition information; According to the global optimization decision of the key parameters and the current driving state of the vehicle, the optimal control sequence of the vehicle is calculated, and the optimal control sequence is used to represent the control mode corresponding to the lowest energy consumption of the vehicle changing from the current driving state to the target state; According to the optimal control sequence, the power system and the brake system of the vehicle are controlled to make the vehicle perform speed regulation operation to reach the target state.

2. The drive control method of a vehicle based on driving assistance according to claim 1, characterized by The road condition information includes front vehicle information, traffic light information and road sign information; The scene semantic analysis is performed by using the road condition information to extract the key parameters corresponding to the road condition information, which comprises the following steps: The front vehicle information, the traffic light information and the road sign information are identified and classified through multi-source data fusion to extract the key parameters, including the distance from the front vehicle, the time of the traffic light and the slope of the road.

3. The drive control method of a vehicle based on drive assist according to claim 2, characterized by The global optimization decision of the key parameters and the current driving state of the vehicle is calculated to obtain the optimal control sequence of the vehicle, which comprises the following steps: Based on the real-time optimization algorithm of dynamic programming or model predictive control, the lowest energy consumption is calculated according to the key parameters and the current driving state; According to the lowest energy consumption and the comprehensive energy consumption of the vehicle, the optimal control sequence is obtained.

4. The drive control method of a vehicle based on drive assist according to claim 3, characterized by The current driving state includes the vehicle speed, the battery state of charge and the battery health state of the vehicle; The real-time optimization algorithm based on dynamic programming or model predictive control is used to calculate the lowest energy consumption according to the key parameters and the current driving state, which comprises the following steps: The pre-constructed cost function is called; The distance from the front vehicle, the time of the traffic light and the slope of the road, as well as the vehicle speed, the battery state of charge and the battery health state of the vehicle are used as parameters to calculate the lowest energy consumption according to the parameters and the cost function.

5. The drive control method of a vehicle based on drive assist according to claim 3, characterized by The global optimization decision of the key parameters and the current driving state of the vehicle is calculated to obtain the optimal control sequence of the vehicle, which comprises the following steps: When the road condition information is that the traffic light is in red light state, the optimal control sequence of the vehicle is determined according to the real-time optimization algorithm, and the optimal control sequence includes the optimal deceleration starting point and curve of the vehicle and the set time; According to the optimal control sequence, the power system and the brake system of the vehicle are controlled, which comprises the following steps: The optimal deceleration starting point and curve are used to control the power system to output zero torque and enter the sliding state; After the vehicle passes through the set time in the sliding state, the brake system is controlled to brake the vehicle.

6. The drive control method of a vehicle based on drive assist according to claim 1 characterized in that, The vehicle also comprises a chassis system with adjustable suspension height, and the vehicle driving control method further comprises the following steps: When the road condition information is that the vehicle is in long distance sliding or downhill state, the suspension of the chassis system is controlled to lower the vehicle body according to the optimal control sequence, and the damping of the suspension is adjusted.

7. A vehicle travel control system based on driving assistance, characterized by The method comprises the following steps: The data acquisition module is configured to acquire road condition information of the vehicle when the vehicle is in the driving assistance mode; The scene semantic parser is configured to perform scene semantic parsing by using the road condition information to extract key parameters corresponding to the road condition information; The global optimization controller is configured to perform global optimization decision according to the key parameters and a current driving state of the vehicle, to calculate an optimal control sequence, and to output a control signal for controlling a power system and a braking system of the vehicle according to the optimal control sequence, so that the vehicle performs speed regulation operation to reach a target state.

8. The drive control system of a vehicle based on drive assist according to claim 1, characterized by, The road condition information includes front vehicle information, traffic light information, and road sign information. The scene semantic parser is further configured to perform scene recognition and classification by using multi-source data fusion on the front vehicle information, the traffic light information, and the road sign information, to extract the key parameters, which include a distance to the front vehicle, a time of the traffic light, and a slope of a road surface. The global optimization controller is further configured to calculate the lowest energy consumption according to the key parameters and the current driving state by using a real-time optimization algorithm based on dynamic programming or model predictive control, and to obtain the optimal control sequence according to the lowest energy consumption and in combination with a comprehensive energy consumption of the vehicle.

9. A vehicle controller characterized by comprising: The vehicle control system comprises: at least one processor; and a memory storing instructions that, when executed by the at least one processor, perform the driving assistance based vehicle driving control method according to any one of claims 1 to 7.

10. A vehicle characterized by comprising: The vehicle control system according to claim 7 or 8, or the vehicle controller according to claim 9.

Images