Commercial vehicle predictive cruise control method and system

Abstract

The invention discloses a predictive cruise control method and system for a commercial vehicle. The method comprises the steps that road gradient information provided by a high-precision map and a target cruise vehicle speed corrected by a deep belief network driving style model are obtained to serve as input parameters; according to the rotating speed and the torque of the engine, a weighted moving average algorithm is adopted to carry out averaging processing on the power value within a short period of time, and the current engine power is generated; on the basis of an automobile power balance equation, according to the current position and the slope value of the front road, the driving power requirement of the next slope section is predicted; and in combination with the efficient output power interval of the engine, the recommended gear and the optimal accelerator opening degree of the next slope section are generated according to different power demand conditions. Therefore, according to the predictive cruise control method for the commercial vehicle, the gears and the accelerator can be adjusted in a self-adaptive mode according to the road gradient and the driving habit, the driving efficiency and the fuel economy are improved, and the individuation of the driving experience is enhanced.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of vehicle control technology, and in particular to a predictive cruise control method and system for a commercial vehicle. Background Art

[0002] The cruise function of commercial vehicles can reduce the driver's fatigue during long-distance driving, allowing the vehicle to run steadily at the set speed, allowing the driver to relax a little. A relatively stable cruising speed can also help reduce the incidence of traffic accidents and avoid common accidents such as rear-end collisions. Long-term stable power output is also more conducive to engine fuel consumption, thereby reducing operating costs. The predictive cruise function is a more advanced cruise function that incorporates information about the road ahead. It automatically adjusts the vehicle speed by analyzing the vehicle's speed, location and road information in real time, and using advanced sensor technology and artificial intelligence algorithms to achieve continuous cruising on more road conditions.

[0003] The existing predictive cruise solutions have the following main disadvantages:

[0004] 1. The calculation and planning of the road ahead often relies on complex calculations and data signal processing, and relies on high-energy consumption algorithms such as dynamic programming and model predictive control to predict the real-time status of the vehicle, which requires high computing power of the controller;

[0005] 2. When calculating, driving economy is often taken as the primary planning goal, and there is less recognition and consideration of driving style, making it difficult to achieve the optimal combination of economy and comfort. Summary of the invention

[0006] The embodiments of the present invention provide a predictive cruise control method and system for commercial vehicles, which can adaptively adjust the gear and throttle according to the road slope and driving habits, improve driving efficiency and fuel economy, and enhance the personalization of driving experience.

[0007] To achieve the above objectives, in a first aspect, the present invention provides a predictive cruise control method for a commercial vehicle, comprising:

[0008] Obtain the road slope information provided by the high-precision map and the target cruising speed corrected by the deep belief network driving style model as input parameters;

[0009] According to the engine speed and torque, the weighted moving average algorithm is used to average the power value in a short period of time to generate the current engine power;

[0010] Based on the vehicle power balance equation, the driving power requirement for the next slope is predicted according to the current position and the slope value of the road ahead;

[0011] Combined with the engine's own efficient output power range and based on different power requirements, the recommended gear and optimal throttle opening for the next slope are generated.

[0012] In one embodiment of the present invention, the averaging process of the power value within a short period of time using a weighted moving average algorithm includes:

[0013] Obtain the power prediction value at t seconds, the power value at t-1 seconds, the weight coefficient, the engine torque at time t, and the engine speed at time t;

[0014] The current engine power is calculated based on the weighted moving average algorithm expression.

[0015] In one embodiment of the present invention, the target cruising speed corrected by the deep belief network driving style model includes:

[0016] For aggressive drivers, a positive speed correction value is given;

[0017] For drivers with normal style, a speed correction value of 0 is assigned;

[0018] For drivers with a peaceful style, a negative speed correction value is given.

[0019] In one embodiment of the present invention, the prediction of the driving power demand of the next slope section based on the vehicle power balance equation according to the current position and the slope value of the road ahead includes:

[0020] Obtain the vehicle's driving power value, transmission efficiency, vehicle weight, friction coefficient, vehicle speed, slope, wind resistance coefficient, windward area and time;

[0021] Based on the vehicle power balance equation, the variable power demand for the next slope section is calculated.

[0022] In one embodiment of the present invention, the generating of the recommended gear and the optimal throttle opening for the next slope section according to different power requirements in combination with the efficient output power range of the engine itself includes:

[0023] Obtaining the efficient output power range corresponding to the optimal fuel consumption range of the engine;

[0024] According to the predicted driving power demand of the next slope section, a target speed value falling within the high-efficiency output power range is determined as a recommended speed for the next slope section;

[0025] Calculate the best gear for the next slope based on the recommended speed and predicted vehicle speed;

[0026] Obtain the speed regulation characteristic MAP, and find the optimal torque value under the condition of known recommended speed;

[0027] The optimal torque value is converted into a throttle opening value to obtain the optimal throttle opening for the next slope section.

[0028] In a second aspect, the present invention provides a predictive cruise control system for commercial vehicles, comprising: an acquisition module, a generation module, a prediction module and a planning module. The acquisition module is used to obtain the road slope information provided by the high-precision map and the target cruising speed corrected by the deep belief network driving style model as input parameters. The generation module is used to average the power value in a short period of time according to the engine speed and torque using a weighted moving average algorithm to generate the current engine power. The prediction module is used to predict the driving power demand of the next slope section based on the vehicle power balance equation according to the current position and the slope value of the road ahead. The planning module is used to combine the engine's own efficient output power range and generate the recommended gear and optimal throttle opening for the next slope section according to different power demand conditions.

[0029] In one embodiment of the present invention, the generation module includes: a first acquisition unit and a first generation unit. The first acquisition unit is used to acquire the power prediction value at t seconds, the power value at t-1 seconds, the weight coefficient, the engine torque at time t, and the engine speed at time t. The first generation unit is used to generate the current engine power according to the weighted moving average algorithm expression.

[0030] In one embodiment of the present invention, the target cruising speed corrected by the deep belief network driving style model includes: for an aggressive style driver, a positive speed correction value is assigned; for a normal style driver, a speed correction value of 0 is assigned; for a moderate style driver, a negative speed correction value is assigned.

[0031] In a third aspect, the present invention provides an electronic device, comprising:

[0032] at least one processor; and

[0033] a memory communicatively coupled to the at least one processor;

[0034] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the commercial vehicle predictive cruise control method as described above.

[0035] In a fourth aspect, the present invention provides a computer-readable storage medium, comprising a computer program and instructions, which, when the computer program or the instructions are executed on a computer, enables the computer to execute the commercial vehicle predictive cruise control method as described above.

[0036] Compared with the prior art, the commercial vehicle predictive cruise control method and system according to the present invention have the following beneficial effects:

[0037] 1. By analyzing the vehicle's speed, location and road information in real time, combined with the engine's efficient output power range, the present invention can intelligently adjust the gear position and throttle opening to keep the engine operating within the optimal fuel consumption range. This can not only reduce unnecessary fuel consumption, but also improve the vehicle's power performance, thereby significantly improving driving efficiency and fuel economy while ensuring driving safety.

[0038] 2. The present invention introduces a deep belief network driving style model, which can personalize the target cruise speed according to the driver's driving style (aggressive, normal, peaceful), so that it can automatically adapt to the driving habits and needs of different drivers, and provide a cruise control experience that is more in line with personal preferences, thereby enhancing driving comfort and satisfaction.

[0039] 3. Existing predictive cruise solutions often rely on complex calculations and data signal processing, such as dynamic programming, model predictive control and other high-energy consumption algorithms. However, the present invention reduces the complexity of calculations and energy consumption requirements by adopting a weighted moving average algorithm and a prediction method based on a physical model, making the system more efficient and energy-saving.

[0040] 4. The present invention uses the road slope information provided by the high-precision map and combines it with the automobile power balance equation to predict the power demand, which can more accurately reflect the actual power demand of the vehicle under different road conditions. At the same time, by combining the engine's efficient output power range for shifting and throttle control, the system can better adapt to the complex and changeable actual road conditions and improve the system's adaptability and robustness. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is a flow chart of a predictive cruise control method for a commercial vehicle in Embodiment 1 of the present invention;

[0042] Figure 2 is a schematic structural diagram of a predictive cruise control system for a commercial vehicle in Embodiment 2 of the present invention;

[0043] Figure 3 It is a schematic diagram of the structure of an electronic device in Embodiment 3 of the present invention;

[0044] Figure 4 is a logic flow diagram of a predictive cruise control method for a commercial vehicle in a specific embodiment of the present invention;

[0045] Figure 5 is a schematic diagram of a high-efficiency output power interval in a specific embodiment of the present invention;

[0046] Figure 6 It is a schematic diagram of the speed regulation characteristic MAP in a specific embodiment of the present invention. DETAILED DESCRIPTION

[0047] The embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the embodiments of the present invention, rather than to limit the embodiments of the present invention. It is also necessary to explain that, for ease of description, only parts related to the embodiments of the present invention are shown in the accompanying drawings, rather than all structures.

[0048] To facilitate understanding, the main implementation concepts of the embodiments of the present invention are first briefly described.

[0049] Embodiment 1, Figure 1 is a flow chart of a predictive cruise control method for a commercial vehicle in Embodiment 1 of the present invention, such as Figure 1 As shown, embodiment 1 provides a predictive cruise control method for a commercial vehicle, including:

[0050] Step S100, obtaining the road slope information provided by the high-precision map and the target cruising speed corrected by the deep belief network driving style model as input parameters;

[0051] Specifically, obtaining the road slope information provided by the high-precision map is one of the important input parameters of the predictive cruise shifting method. The high-precision map can provide detailed information such as the slope, curve radius, speed limit, etc. of the road ahead of the vehicle through precise three-dimensional modeling and mapping technology. For example, sensors such as laser radar and inertial navigation system are used to collect road data, and high-precision maps are generated through point cloud stitching, feature extraction and other algorithms. The system can obtain the road slope change within a certain mileage of the vehicle in the future from the map, such as the slope data within 500 meters ahead. The road is divided into several slope sections, each of which is characterized by the starting mileage, slope section length and slope value. The system obtains the current position information of the vehicle in real time, and queries the slope information of the matching road section ahead from the map database as one of the input conditions for subsequent power demand prediction and shifting strategy formulation. The deep belief network driving style model is used to correct the target cruise speed to meet the personalized needs of different drivers. This model is a driving behavior recognition algorithm based on deep learning. By learning the long-term driving data of the driver, the mapping relationship between driving style characteristics and vehicle speed preferences is established. The input features of the model can include the statistical feature values ​​of signals such as the accelerator pedal opening, brake pedal opening, steering wheel angle, etc., such as mean, standard deviation, rate of change, etc. By training massive user data on the cloud service platform, a driving style recognition model with strong generalization ability is obtained. When the model is applied, the system collects the driver's driving data for a period of time, extracts features and inputs them into the model for style recognition, and obtains a style grade value. The style grade value is weighted and summed with the preset target cruise speed value to obtain a personalized and corrected target speed. For example, a driver's style grade is 0.8, and he prefers a faster speed. The original target speed was set to 80km / h, which was increased to 85km / h after correction. The corrected speed is used as the input value for subsequent calculations, organically combining the driver's subjective preference factors with objective vehicle operating conditions factors to achieve personalized and intelligent predictive cruise control.

[0052] Step S200, according to the engine speed and torque, a weighted moving average algorithm is used to average the power value in a short period of time to generate the current engine power; wherein the weighted moving average algorithm is specifically:

[0053]

[0054] Specifically, by obtaining the engine speed n in real time (t) and torque T tq(t) , and combined with the power value P at the previous moment t-1 and the predicted value at the current moment are processed using the weighted moving average algorithm to generate the engine power P at the current moment (t). Among them, β is the weight coefficient, which is used to adjust the weight distribution of historical power values ​​and current predicted power values ​​in the calculation. This method not only takes into account the current working status of the engine, but also introduces historical data to make the power calculation smoother and more stable, reducing the error caused by instantaneous fluctuations. Specifically, by taking the weighted average of the predicted power value at the current moment (calculated based on speed and torque) and the actual power value at the previous moment, it can more accurately reflect the dynamic changes of engine power, and provide reliable basic data for the subsequent power demand prediction and shift strategy formulation based on the vehicle power balance equation. The application of this algorithm enables commercial vehicles to achieve more accurate and economical cruise control under different road conditions and driving styles, thereby improving driving efficiency, fuel economy and personalized driving experience.

[0055] For example, a commercial vehicle is driving, and its engine speed and torque data are measured in real time by sensors. At the current time t = 10 seconds, the engine power P at t = 10 seconds needs to be calculated. 10 First, obtain the following parameters: power prediction value P' at t = 10 seconds (10) , this value can be obtained by using certain prediction algorithms (such as Kalman filtering, neural networks, etc.) based on the power value at the previous moment and the speed, torque and other parameters at the current moment. Assume that P '(10) The actual power value P at t = 9 seconds is 120kW. (9) , this value can be read directly from the engine control unit (ECU). Assume P (9) The weight coefficient β is used to balance the weight of the current prediction value and the actual value at the previous moment, and is usually between 0 and 1. Assume that β is 0.6. The engine torque T at t = 10 seconds tq(10) , assuming it is 300 Nm. Engine speed n at t = 10 seconds (10) , assuming it is 2000 rpm. Next, use the weighted moving average algorithm formula to calculate P 10 :

[0056]

[0057] Substitute the known values:

[0058]

[0059] The calculation results are:

[0060]

[0061] P 10 =72+0.4*62.83=97.13kW

[0062] Therefore, through the weighted moving average algorithm, the engine power at t = 10 seconds is calculated to be 97.13kW. This value combines the predicted power value at the current moment and the actual power value at the previous moment. By adjusting the weight coefficient, it not only takes into account the dynamic changes of the current state, but also maintains the stability and reliability of the calculation results, providing an important reference for the subsequent power demand prediction and shift strategy formulation based on the vehicle power balance equation.

[0063] Step S300, based on the vehicle power balance equation, according to the current position and the slope value of the road ahead, predict the driving power demand of the next slope section; wherein the vehicle power balance equation is:

[0064]

[0065] Specifically, this step uses the vehicle power balance equation to comprehensively consider the vehicle's current position and the slope information of the road ahead, as well as other relevant parameters (such as transmission efficiency, vehicle weight, friction coefficient, vehicle speed, wind resistance coefficient, frontal area, etc.) to predict the power required for the vehicle to travel on the next slope. The vehicle power balance equation is a complex physical model that accurately describes the various resistances that need to be overcome during the vehicle's driving process, including rolling resistance, air resistance, slope resistance, and acceleration resistance. Through this step, the system can understand the vehicle's power requirements under different road conditions in advance, providing an important reference for subsequent shifting strategies and throttle control. This not only ensures that the vehicle can travel in the optimal state under different road conditions, but also significantly improves driving efficiency and fuel economy, while improving driving comfort and safety. Therefore, power demand prediction based on the vehicle power balance equation is the core link in the predictive cruise control method for commercial vehicles, which provides a solid foundation for realizing intelligent and personalized cruise control.

[0066] For example, a fully loaded heavy truck is driving on the highway. The current speed is 70 kilometers per hour, the transmission efficiency is 0.85, the vehicle weight is 45 tons, the friction coefficient is 0.02, the drag coefficient is 0.5, and the frontal area is 7 square meters. Through the high-precision map, the system obtains the slope information of the road within 500 meters ahead. The slope at the current position is 0%, and there is an uphill section starting 200 meters ahead with a slope of 3%. Based on this information, the system begins to use the car power balance equation to predict the driving power demand of the next slope section. First, substitute the known parameters into the equation:

[0067]

[0068] Among them, η T is the transmission efficiency of 0.85, G is the vehicle weight of 45,000 kg, f is the friction coefficient of 0.02, u ais the vehicle speed of 70 kilometers per hour (i.e. 70 / 3.6 meters per second), i is the slope of 0% (current position) and 3% (next slope section), C D is the drag coefficient of 0.5, A is the frontal area of ​​7 square meters, δ is the rotation mass conversion factor (usually 1.05 is used to simplify the calculation), is the acceleration (assuming it is currently traveling at a constant speed, so it is 0).

[0069] For current position (0% slope):

[0070]

[0071] After simplifying the calculation, the driving power demand at the current position is mainly determined by rolling resistance and air resistance.

[0072] For the next slope (3% slope):

[0073]

[0074] Note the increase in the slope resistance term, as going uphill requires overcoming the component of gravity along the slope.

[0075] Through calculation, the system can predict that in the next slope section, due to the increase in slope, the vehicle needs to provide more power to overcome the component of gravity along the slope to maintain the current speed. Assume that the calculation result is that the driving power demand at the current position is 150 kilowatts, and the driving power demand for the next slope section increases to 170 kilowatts. This means that in order to maintain a speed of 70 kilometers per hour uphill, the engine needs to output more power.

[0076] Based on this prediction result, the system can further combine the engine's efficient output power range to intelligently plan the appropriate gear and throttle opening to ensure that the vehicle can pass the uphill section smoothly and economically. This not only improves driving efficiency, but also optimizes fuel economy, while enhancing driving comfort and safety.

[0077] Step S400, combining the efficient output power range of the engine itself and generating the recommended gear and optimal throttle opening for the next slope according to different power requirements;

[0078] Specifically, the system first obtains the efficient output power range corresponding to the engine's optimal fuel consumption range. This range is obtained through experimental data and characteristic curve analysis, representing the area where the engine can convert fuel into power with the highest efficiency and produce less emissions under different speed and torque combinations. Then, the system will determine a target speed value that falls within this efficient output power range as the recommended speed based on the previously predicted power demand for the next slope. This recommended speed is designed to meet the power demand while achieving the best fuel economy. With the recommended speed, the system will calculate the most suitable gear at the recommended speed by querying the vehicle's transmission ratio data in combination with the current vehicle speed. For example, if the current vehicle speed is 60 km / h and the recommended speed is 2500 rpm, the system may find that 4th gear is the gear closest to this speed, so it is determined as the best gear. Finally, the system will query the engine's speed regulation characteristic MAP (torque characteristic map), which is a data table that pre-stores the torque output characteristics of the engine at different speeds and throttle openings. Under the condition of knowing the recommended speed, the system can find the optimal throttle opening required to output the target torque. For example, if the recommended speed is 2500 rpm, and in order to meet the power requirement, a torque of 180 Nm needs to be output, the system will find the corresponding throttle opening value, such as 40%, through the MAP table. In this way, the system generates the recommended gear (such as 4th gear) and the optimal throttle opening (such as 40%) for the next slope, thereby achieving precise control of the engine power output, meeting the power requirement and optimizing fuel economy.

[0079] For example, a heavy truck is driving on a road with multiple slopes. The current speed is 70 kilometers per hour, and the road ahead is about to enter an uphill section with a slope of 5%. The system has obtained the slope information ahead through the high-precision map, and predicted through the vehicle power balance equation that in order to maintain the current speed uphill, the vehicle needs to provide about 220 kilowatts of power. Next, the system plans the shift strategy and throttle control in combination with the engine's own efficient output power range. Assume that the engine's optimal fuel consumption range corresponds to an efficient output power range of 2000-2500 rpm and a torque output of 180-220 Nm. The system calculated that when the speed is 2300 rpm and the torque output is 200 Nm, the engine can output 220 kilowatts of power, and this working point falls exactly within the efficient output power range. Therefore, the system determines 2300 rpm as the recommended speed for the next slope. Then, based on the current vehicle speed and the recommended speed, combined with the vehicle's transmission ratio data, the system calculates that the most suitable gear is 4th gear at 2300 rpm. Because at the current speed, 4th gear can make the engine speed closest to the recommended 2300 rpm. Next, the system queries the engine's speed regulation characteristic MAP (torque characteristic map) and finds the throttle opening value corresponding to the output of 200 Nm of torque under the condition of knowing the recommended speed of 2300 rpm. Assume that the MAP table shows that at 2300 rpm, the throttle opening required to output 200 Nm of torque is 55%. Therefore, the system finally generates the recommended gear for the next slope as 4th gear, and the optimal throttle opening is 55%. In this way, when the truck is going uphill, it can not only provide enough power to maintain the vehicle speed, but also ensure that the engine works in the efficient output power range, thereby achieving the best fuel economy. At the same time, by planning the gear shift and throttle control in advance, the system can also ensure the smoothness of the gear shift process and improve driving comfort.

[0080] In this embodiment, the weighted moving average algorithm is used to average the power value within a short period of time, including: obtaining the power prediction value at t seconds, the power value at t-1 seconds, the weight coefficient, the engine torque at time t, and the engine speed at time t; according to the weighted moving average algorithm expression, the current engine power is calculated;

[0081] Specifically, a series of key parameters are first obtained, including the power prediction value at t seconds, the actual power value at t-1 seconds, a coefficient for balancing the weight of new and old data (i.e., weight coefficient), the engine torque at time t, and the engine speed at time t. These parameters are the basis for calculating the current engine power. The power prediction value is usually based on the power value at the previous moment, the engine torque and speed at the current moment, and possible other dynamic factors (such as vehicle acceleration, road conditions, etc.), and is obtained through a certain prediction algorithm (such as Kalman filtering, neural network, etc.). This prediction value reflects the system's instant estimate of the current engine power. The actual power value at t-1 seconds is the actual power output value of the previous moment directly read by the engine control unit (ECU). This value represents the true performance of the engine at a certain moment in the past. The weight coefficient is a constant with a value between 0 and 1, which is used to adjust the relative importance of the power prediction value and the actual power value in calculating the current power average value. When the weight coefficient is large, the predicted value has a greater impact on the final result; conversely, when the weight coefficient is small, the actual value has a more significant impact. By adjusting this coefficient, the system can find a balance between the dynamics of the predicted value and the stability of the actual value. The engine torque and speed at time t are measured in real time by sensors installed on the engine. These two parameters directly reflect the current working state of the engine and are indispensable data for calculating power. After obtaining these parameters, the system will calculate according to the expression of the weighted moving average algorithm. This expression weights the power prediction value and the actual power value according to the weight coefficient to obtain a smoother and more stable current engine power value. This value not only takes into account the current real-time state of the engine, but also incorporates the influence of historical data, making the calculation result more reliable and accurate. The advantage of this method is that it can effectively integrate the predicted value and the actual value, retaining the flexibility and foresight of the predicted value while taking into account the stability and reliability of the actual value. At the same time, by adjusting the weight coefficient, the system can flexibly adjust the proportion of the predicted value and the actual value in the calculation according to different application scenarios and needs, thereby achieving more accurate and efficient engine power control. In the predictive cruise control of commercial vehicles, the application of this method can significantly improve driving efficiency and fuel economy, while enhancing the personalization of the driving experience.

[0082] In this embodiment, the target cruising speed corrected by the deep belief network driving style model includes: for an aggressive style driver, a positive speed correction value is assigned; for a normal style driver, a speed correction value of 0 is assigned; for a moderate style driver, a negative speed correction value is assigned;

[0083] Specifically, for aggressive drivers, a positive speed correction value is assigned, which means that the system will appropriately increase the target cruising speed. The purpose of this is to meet the aggressive driver's pursuit of power performance. By increasing the target speed, the engine will run at a higher speed and load, thereby providing stronger acceleration performance and climbing ability. At the same time, the system will adjust the shift strategy accordingly, delay the timing of upshifting, and keep the engine in a higher speed range to ensure sufficient power output. For example, the original upshift to a higher gear at 80 kilometers per hour may now be postponed to 85 kilometers. In this way, aggressive drivers can get a more enjoyable driving experience. For normal drivers, a speed correction value of 0 is assigned, that is, the original target cruising speed remains unchanged. This shows that the system believes that the current speed setting can well balance power and economy, and no additional adjustment is needed. Normal driving style usually pursues a smooth and comfortable driving experience, and also hopes to obtain better fuel economy. Therefore, maintaining the default cruising speed and shifting logic can achieve a more ideal comprehensive performance while meeting the driver's expectations. For drivers with a calm style, a negative speed correction value is given, that is, the target cruising speed is appropriately reduced. This approach is intended to further optimize fuel economy while providing a more comfortable and relaxed driving experience. By reducing the speed, the engine can run at a lower speed and load, thereby reducing fuel consumption. At the same time, the system will adjust the shift strategy accordingly, shifting to a higher gear in advance, so that the engine can be maintained in a low-speed and high-torque range as much as possible to improve fuel efficiency. For example, the original shift to a higher gear at 90 kilometers per hour may now be shifted to 85 kilometers in advance. In this way, calm drivers can enjoy a more comfortable and pleasant driving experience, while also achieving better fuel economy performance. In general, by introducing a deep belief network driving style model and making targeted corrections to the target cruising speed according to different styles, the system can better adapt to the personalized needs of drivers. Aggressive drivers can get stronger power performance, normal drivers can take into account both power and economy, and calm drivers can enjoy a more comfortable and fuel-efficient driving experience. This personalized adjustment strategy can significantly improve driver satisfaction and driving experience, while also optimizing vehicle performance as a whole and achieving the best balance between power, economy and comfort.

[0084] In this embodiment, the prediction of the driving power demand of the next slope section based on the vehicle power balance equation according to the current position and the slope value of the road ahead includes: obtaining the vehicle driving power value, transmission efficiency, vehicle weight, friction coefficient, vehicle speed, slope, wind resistance coefficient, frontal area and time; calculating the variable power demand of the next slope section according to the vehicle power balance equation;

[0085] Specifically, obtaining parameters such as the vehicle's driving power value, transmission efficiency, vehicle weight, friction coefficient, vehicle speed, slope, wind resistance coefficient, windward area and time is the basis for predicting the driving power demand of the next slope section. Among them, the vehicle's driving power value reflects the power required for the vehicle to overcome the driving resistance, which is closely related to factors such as vehicle speed and slope. For example, on an uphill section, the vehicle needs to overcome greater gravitational potential energy, so the driving power value will increase. The transmission efficiency indicates the efficiency of the engine output power transmitted to the wheels, which is affected by the transmission system such as the gearbox and differential. The vehicle weight directly affects the inertia and rolling resistance of the vehicle. Under heavy load conditions, the power demand of the vehicle will increase significantly. The friction coefficient depends on the road conditions. On slippery, muddy and other low-adhesion roads, the friction coefficient decreases, the vehicle is prone to wheel slippage, and the driving force transmitted to the ground is reduced. The square of the vehicle speed is proportional to the air resistance. The higher the speed, the greater the wind resistance the vehicle encounters, and more power is required to overcome it. Slope information can be obtained from high-precision maps. The greater the slope, the greater the power required for the vehicle to climb. The drag coefficient and the frontal area jointly determine the air resistance of the car. The streamlined body has a smaller drag coefficient and the car's driving resistance is reduced. In addition, since the power demand of the car is time-varying, it is necessary to obtain time information in order to predict the power changes in the future. After obtaining the above parameters, the car power balance equation can be used to calculate the variable power demand of the next slope. This equation takes into account the various resistances to the car, including rolling resistance, air resistance, slope resistance, etc., as well as the acceleration resistance of the car. For example, assume that a fully loaded heavy truck weighs 40 tons, has a speed of 60 kilometers per hour, a transmission efficiency of 0.8, a friction coefficient of 0.018, a drag coefficient of 0.6, a frontal area of ​​6 square meters, and a current road slope of 5%. It is predicted that the slope will remain unchanged in the next 10 seconds. Substituting the above data into the power balance equation, it can be concluded that to maintain the current speed, the car needs to provide a power of about 200 kilowatts. If the engine is working in the economic range at this time, and its output power is 180 kilowatts, it means that the current power output is insufficient, and it is necessary to increase the engine speed and throttle opening, or shift to a low-speed gear and reduce the transmission ratio to meet the power demand. On the contrary, if the calculated power demand is less than the current output power of the engine, you can consider increasing the gear and reducing the throttle opening to improve fuel economy. In summary, based on the automobile power balance equation, it is necessary to comprehensively consider vehicle parameters, road conditions, environmental conditions and other factors to predict the driving power demand of the next slope. By obtaining these parameters in real time and substituting them into the power balance equation for calculation, the power demand changes in the future can be predicted, providing a basis for intelligent shifting and throttle control, while meeting the power demand, taking into account fuel economy and emission performance, and realizing the comprehensive performance optimization of the vehicle. This predictive control method based on the physical model has better adaptability and robustness than the empirical model, and can cope with complex and changeable actual road conditions.At the same time, the introduction of driver behavior pattern recognition can further improve the prediction accuracy, make the control strategy closer to the driver's subjective intention, and enhance the driving experience.

[0086] In this embodiment, the generating of the recommended gear and the optimal throttle opening for the next slope section according to different power requirements in combination with the efficient output power range of the engine itself includes:

[0087] Obtaining the efficient output power range corresponding to the optimal fuel consumption range of the engine;

[0088] According to the predicted driving power demand of the next slope section, a target speed value falling within the high-efficiency output power range is determined as a recommended speed for the next slope section;

[0089] Calculate the best gear for the next slope based on the recommended speed and predicted vehicle speed;

[0090] Obtain the speed regulation characteristic MAP, and find the optimal torque value under the condition of known recommended speed;

[0091] Converting the optimal torque value into a throttle opening value to obtain an optimal throttle opening for the next slope section;

[0092] Specifically, the efficient output power range of the engine corresponds to its optimal fuel consumption range, which is not a fixed value, but an area. The determination of this area is usually obtained through the experimental data and characteristic curve analysis of the engine. The fuel efficiency of the engine is different at different speeds and torque outputs. Under certain specific speed and torque combinations, the engine can convert fuel into power with the highest efficiency while producing less emissions. This most efficient area is the efficient output power range. For example, a 2.0L naturally aspirated engine may be in its optimal fuel consumption range, that is, the efficient output power range, when the speed is 2000-3000 rpm and the torque output is 150-200 Nm. The reason for using this range as a reference is that keeping the engine working in this area can minimize fuel consumption and improve fuel economy. Assume that the predicted driving power demand for the next slope is 30 kilowatts. By querying the engine characteristic curve, it is found that to achieve a power output of 30 kilowatts, the engine can achieve it under multiple speed and torque combinations. However, in order to ensure fuel economy, it is necessary to select a speed value that falls within the efficient output power range. Assume that the engine's efficient output power range corresponds to a speed range of 2200 rpm to 2800 rpm. After calculation, when the engine speed is 2500 rpm, the torque output is 180 Nm, which just meets the power requirement of 30 kW, and the speed value is within the efficient output power range. Therefore, 2500 rpm is determined as the recommended speed for the next slope. This process is not a simple matching, but through the algorithm, under the premise of meeting the power demand, try to choose the speed within the efficient output range to achieve the energy saving goal. After determining the recommended speed, it is also necessary to calculate the best gear in combination with the predicted vehicle speed. For example, the current vehicle speed is 60 km / h, and the recommended speed is 2500 rpm. By querying the vehicle's transmission ratio data, you can know the corresponding relationship between the engine speed and the vehicle speed at different gears. Assume that the engine speed is 3000 rpm in the 3rd gear at a speed of 60 km / h, and the engine speed is 2200 rpm in the 4th gear at a speed of 60 km / h, and the speed is lower in the 5th gear. Considering that the recommended speed is 2500 rpm, 4th gear is closest, so the best gear is determined to be 4th gear. The logic of shifting is to try to choose a high gear while meeting the recommended speed to reduce the engine speed and thus reduce fuel consumption. If the calculation finds that the current gear already meets the speed requirements, there is no need to shift gears and just keep the current gear. The speed control characteristic MAP, also known as the engine torque characteristic map, is a pre-stored data table that describes the torque output characteristics of the engine at different speeds and throttle openings. This MAP table is usually calibrated using a large amount of experimental data, and it reflects the performance limit of the engine.Under the condition of known recommended speed, by querying this MAP table, we can find the maximum torque that the engine can output at this speed, and the throttle opening corresponding to the target torque output at this speed. For example, if the recommended speed is 2500 rpm, by querying the speed regulation characteristic MAP, it is found that at 2500 rpm, the maximum torque of the engine is 200 Nm. In order to meet the power demand of 30 kW, only 180 Nm of torque is needed. According to the MAP table, 180 Nm of torque corresponds to a specific throttle opening value, such as 40%. The throttle opening value is actually the opening size of the valve that controls the engine's intake volume. The larger the throttle opening, the more air enters the cylinder, the more fuel injection volume increases accordingly, and the greater the torque output by the engine. After finding the optimal torque value, the process of converting it into the throttle opening value is actually a control process, the purpose of which is to make the engine work according to the calculated optimal torque output. For example, as mentioned earlier, when 180 Nm of torque is required, the throttle opening is 40% by querying the MAP table. Then, the control system will adjust the throttle opening to 40%, thereby ensuring that the engine outputs 180 Nm of torque. This process ensures that the engine operates in the most economical way and achieves the best fuel consumption performance while meeting the power requirements. At the same time, by controlling the throttle opening, precise control of the engine power output can be achieved to avoid unnecessary fuel consumption.

[0093] In this embodiment, the deep belief network driving style model is deployed on a cloud service platform, and a wireless communication connection is established with the vehicle through an on-board communication module, thereby realizing personalized recognition of the user's driving style.

[0094] In practical applications, the present invention uses the road slope information provided by the high-precision map and the target cruising speed corrected by the deep belief network driving style model as input parameters. Based on the theoretical foundation of the automobile power balance equation, the system can predict the driving power demand of the next slope section, and combine the engine's own efficient output power range to intelligently plan the recommended gear and optimal throttle opening for the next slope section. This not only realizes continuous cruising driving planning in multiple road scenarios, but also fully meets the user's high requirements for economy and driving comfort. Figure 4 As shown, each stage of the present invention is described in detail as follows:

[0095] 1. Using the engine speed and torque, use the weighted moving average algorithm to average the power value in a short period of time to obtain the current engine power. The weighted moving average algorithm expression is:

[0096]

[0097] P t is the power prediction value at t seconds, P t-1is the power value at t-1 seconds, β is the weight coefficient, T tq(t) is the engine torque at time t, n (t) is the engine speed at time t.

[0098] 2. Based on the deep belief network driving style recognition model, the speed correction value of the vehicle overhill is assigned, with positive values ​​for aggressive style, 0 for normal style, and negative values ​​for peaceful style.

[0099] 3. Obtain the current position and the slope value of the road ahead from the high-precision map terminal, and calculate the variable power demand of the next slope based on the vehicle power balance equation. Vehicle power balance equation:

[0100]

[0101] P e is the driving power value of the car, η T is the transmission efficiency, G is the vehicle weight, f is the friction coefficient, u a is the vehicle speed, i is the slope, C D is the drag coefficient, A is the frontal area, and t is the time.

[0102] 4. After predicting the driving power demand of the next slope section, the engine's own high-efficiency output power range is combined with different power demand conditions to intelligently plan the speed value of the next slope section. The high-efficiency output power range is the optimal fuel consumption range. The dark area between the two horizontal lines in the figure is the area with the lowest specific fuel consumption. Figure 5 shown.

[0103] 5. Based on the recommended speed and vehicle speed, the best gear for the next slope can be calculated. By searching the speed control characteristic MAP, the optimal torque value can be found under the condition of known speed. The engine will automatically convert the current speed and torque requirements into the throttle opening value to obtain the optimal throttle opening for the next slope. The speed control characteristic MAP diagram is shown in the figure below. Figure 6 shown.

[0104] Embodiment 2: Figure 2 is a schematic diagram of the structure of a predictive cruise control system for a commercial vehicle in Embodiment 2 of the present invention, such as Figure 2As shown, embodiment 2 provides a predictive cruise control system for commercial vehicles, including: an acquisition module 201, a generation module 202, a prediction module 203 and a planning module 204. The acquisition module 201 is used to obtain the road slope information provided by the high-precision map and the target cruising speed corrected by the deep belief network driving style model as input parameters. The generation module 202 is used to average the power value in a short period of time according to the engine speed and torque using a weighted moving average algorithm to generate the current engine power. The prediction module 203 is used to predict the driving power demand of the next slope section based on the vehicle power balance equation according to the current position and the slope value of the road ahead. The planning module 204 is used to combine the engine's own efficient output power range and generate the recommended gear and optimal throttle opening for the next slope section according to different power demand conditions.

[0105] In this embodiment, the generation module 202 includes: a first acquisition unit and a first generation unit. The first acquisition unit is used to acquire the power prediction value at t seconds, the power value at t-1 seconds, the weight coefficient, the engine torque at time t, and the engine speed at time t. The first generation unit is used to generate the current engine power according to the weighted moving average algorithm expression.

[0106] In this embodiment, the target cruising speed corrected by the deep belief network driving style model includes: for an aggressive style driver, a positive speed correction value is assigned; for a normal style driver, a speed correction value of 0 is assigned; for a moderate style driver, a negative speed correction value is assigned.

[0107] In this embodiment, the prediction module 203 includes: a second acquisition unit and a first calculation unit. The second acquisition unit is used to obtain the vehicle driving power value, transmission efficiency, vehicle weight, friction coefficient, vehicle speed, slope, wind resistance coefficient, frontal area and time. The first calculation unit is used to calculate the variable power demand of the next slope section according to the vehicle power balance equation.

[0108] In this embodiment, the planning module 204 includes: a third acquisition unit, a determination unit, a second calculation unit, a search unit and a second generation unit. The third acquisition unit is used to obtain the high-efficiency output power interval corresponding to the optimal fuel consumption interval of the engine. The determination unit is used to determine the target speed value falling within the high-efficiency output power interval as the recommended speed for the next slope section based on the predicted driving power demand for the next slope section. The second calculation unit is used to calculate the optimal gear for the next slope section based on the recommended speed and the predicted vehicle speed. The search unit is used to obtain the speed regulation characteristic MAP, and find the optimal torque value under the condition of knowing the recommended speed. The second generation unit is used to convert the optimal torque value into a throttle opening value to generate the optimal throttle opening for the next slope section.

[0109] In this embodiment, the deep belief network driving style model is deployed on a cloud service platform, and a wireless communication connection is established with the vehicle through an on-board communication module, thereby realizing personalized recognition of the user's driving style.

[0110] The various variations and specific examples of the commercial vehicle predictive cruise control method provided in Example 1 are also applicable to the commercial vehicle predictive cruise control system provided in this embodiment. Through the above detailed description of a commercial vehicle predictive cruise control method, those skilled in the art can clearly know the implementation method of a commercial vehicle predictive cruise control system in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here.

[0111] Embodiment 3, Figure 3 is a schematic diagram of the structure of an electronic device in Embodiment 3 of the present invention, such as Figure 3 As shown, the third embodiment further provides an electronic device 300 , which may include: a processor 301 and a memory 302 .

[0112] The memory 302 is used to store programs; the memory 302 may include volatile memory (English: volatile memory), such as random-access memory (English: random-access memory, abbreviated: RAM), such as static random-access memory (English: static random-access memory, abbreviated: SRAM), double data rate synchronous dynamic random access memory (English: Double Data Rate Synchronous Dynamic Random Access Memory, abbreviated: DDR SDRAM), etc.; the memory may also include non-volatile memory (English: non-volatile memory), such as flash memory (English: flash memory). The memory 302 is used to store computer programs (such as applications, functional modules, etc. that implement the above method), computer instructions, etc., and the above computer programs, computer instructions, etc. can be partitioned and stored in one or more memories 302. And the above computer programs, computer instructions, data, etc. can be called by the processor 301.

[0113] The above-mentioned computer programs, computer instructions, etc. may be stored in partitions in one or more memories 302 . And the above-mentioned computer programs, computer instructions, etc. may be called by the processor 301 .

[0114] The processor 301 is used to execute the computer program stored in the memory 302 to implement each step of the method involved in the above embodiment.

[0115] For details, please refer to the relevant description in the previous method embodiment.

[0116] The processor 301 and the memory 302 may be independent structures or integrated structures. When the processor 301 and the memory 302 are independent structures, the memory 302 and the processor 301 may be coupled and connected via a bus 303 .

[0117] The electronic device of this embodiment can execute the technical solution in the above method, and its specific implementation process and technical principle are the same, which will not be repeated here.

[0118] Embodiment 4: Embodiment 4 also provides a computer-readable storage medium, including a computer program and instructions. When the computer program or instructions are executed on a computer, the computer executes the commercial vehicle predictive cruise control method of any embodiment of the present invention.

[0119] Computer-readable storage media include: USB flash drives, mobile hard disks, ROM, RAM, magnetic disks, or optical disks, and other media that can store program codes.

[0120] This embodiment also provides a computer program product, which includes: a computer program, which is stored in a readable storage medium. At least one processor of an electronic device can read the computer program from the readable storage medium, and at least one processor executes the computer program so that the electronic device executes the solution provided by any of the above embodiments.

[0121] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the disclosure of the present invention can be performed in parallel, sequentially or in different orders, as long as the desired results of the technical solution disclosed in the present invention can be achieved, and this document does not limit this.

[0122] In summary, the commercial vehicle predictive cruise control method and system of the present invention have the following beneficial effects:

[0123] 1. By analyzing the vehicle's speed, location and road information in real time, combined with the engine's efficient output power range, the present invention can intelligently adjust the gear position and throttle opening to keep the engine operating within the optimal fuel consumption range. This can not only reduce unnecessary fuel consumption, but also improve the vehicle's power performance, thereby significantly improving driving efficiency and fuel economy while ensuring driving safety.

[0124] 2. The present invention introduces a deep belief network driving style model, which can personalize the target cruise speed according to the driver's driving style (aggressive, normal, peaceful), so that it can automatically adapt to the driving habits and needs of different drivers, and provide a cruise control experience that is more in line with personal preferences, thereby enhancing driving comfort and satisfaction.

[0125] 3. Existing predictive cruise solutions often rely on complex calculations and data signal processing, such as dynamic programming, model predictive control and other high-energy consumption algorithms. However, the present invention reduces the complexity of calculations and energy consumption requirements by adopting a weighted moving average algorithm and a prediction method based on a physical model, making the system more efficient and energy-saving.

[0126] 4. The present invention uses the road slope information provided by the high-precision map and combines it with the automobile power balance equation to predict the power demand, which can more accurately reflect the actual power demand of the vehicle under different road conditions. At the same time, by combining the engine's efficient output power range for shifting and throttle control, the system can better adapt to the complex and changeable actual road conditions and improve the system's adaptability and robustness.

[0127] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A predictive cruise control method for a commercial vehicle, characterized in that: include: Obtain the road slope information provided by the high-precision map and the target cruising speed corrected by the deep belief network driving style model as input parameters; According to the engine speed and torque, the weighted moving average algorithm is used to average the power value in a short period of time to generate the current engine power; Based on the vehicle power balance equation, the driving power requirement for the next slope is predicted according to the current position and the slope value of the road ahead; Combined with the engine's own efficient output power range and based on different power requirements, the recommended gear and optimal throttle opening for the next slope are generated.

2. The commercial vehicle predictive cruise control method according to claim 1, characterized in that: The method of using a weighted moving average algorithm to average the power value within a short period of time includes: Obtain the power prediction value at t seconds, the power value at t-1 seconds, the weight coefficient, the engine torque at time t, and the engine speed at time t; The current engine power is calculated based on the weighted moving average algorithm expression.

3. The commercial vehicle predictive cruise control method according to claim 1, characterized in that: The target cruising speed corrected by the deep belief network driving style model includes: For aggressive drivers, a positive speed correction value is given; For drivers with normal style, a speed correction value of 0 is assigned; For drivers with a peaceful style, a negative speed correction value is given.

4. The commercial vehicle predictive cruise control method according to claim 1, characterized in that: The method of predicting the driving power demand of the next slope section based on the vehicle power balance equation and the current position and the slope value of the road ahead includes: Obtain the vehicle's driving power value, transmission efficiency, vehicle weight, friction coefficient, vehicle speed, slope, wind resistance coefficient, windward area and time; Based on the vehicle power balance equation, the variable power demand for the next slope section is calculated.

5. The commercial vehicle predictive cruise control method according to claim 1, characterized in that: The generating of the recommended gear and the optimal throttle opening for the next slope section according to different power requirements in combination with the efficient output power range of the engine itself includes: Obtaining the efficient output power range corresponding to the optimal fuel consumption range of the engine; According to the predicted driving power demand of the next slope section, a target speed value falling within the high-efficiency output power range is determined as a recommended speed for the next slope section; Calculate the best gear for the next slope based on the recommended speed and predicted vehicle speed; Obtain the speed regulation characteristic MAP, and find the optimal torque value under the condition of known recommended speed; The optimal torque value is converted into a throttle opening value to obtain the optimal throttle opening for the next slope section.

6. A predictive cruise control system for a commercial vehicle, characterized in that: include: An acquisition module is used to obtain the road slope information provided by the high-precision map and the target cruising speed corrected by the deep belief network driving style model as input parameters; A generation module is used to average the power value within a short period of time according to the engine speed and torque using a weighted moving average algorithm to generate the current engine power; A prediction module is used to predict the driving power demand of the next slope section based on the vehicle power balance equation according to the current position and the slope value of the road ahead; The planning module is used to combine the engine's own efficient output power range and generate the recommended gear and optimal throttle opening for the next slope according to different power requirements.

7. The predictive cruise control system for commercial vehicles according to claim 6, characterized in that: The generation module comprises: The first acquisition unit is used to acquire the power prediction value at t seconds, the power value at t-1 seconds, the weight coefficient, the engine torque at time t, and the engine speed at time t; The first generating unit is used to generate the current engine power according to the weighted moving average algorithm expression.

8. The predictive cruise control system for commercial vehicles according to claim 6, characterized in that: The target cruising speed corrected by the deep belief network driving style model includes: For aggressive drivers, a positive speed correction value is given; For drivers with normal style, a speed correction value of 0 is assigned; For drivers with a peaceful style, a negative speed correction value is given.

9. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively coupled to the at least one processor; The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the commercial vehicle predictive cruise control method according to any one of claims 1 to 5.

10. A computer-readable storage medium, characterized in that: The method comprises a computer program and instructions. When the computer program or the instructions are executed on a computer, the computer is enabled to execute the commercial vehicle predictive cruise control method according to any one of claims 1 to 5.

Images