A cloud-based speed planning system for new energy commercial vehicles based on artificial intelligence

Abstract

This invention relates to the field of intelligent connected vehicle technology, specifically to an artificial intelligence-based cloud-based speed planning system for new energy commercial vehicles. The system uses a data acquisition module to collect data from the vehicle, including route planning, vehicle weight, continuous driving time, battery SOC, vehicle configuration information, system time, and weather. It also collects road information from the cloud, such as gradient, road grade, curve curvature, number of lanes, and speed limits. After being structured according to certain rules by a data processing module, the data is input into an algorithm model generated by a cloud database and artificial intelligence algorithms. This results in a speed curve on the planned road that balances comfort, economy, and transportation efficiency. As samples accumulate, the quality of the database and the accuracy of the algorithm model are continuously optimized, ultimately reducing the driver's workload, lowering vehicle energy consumption, and improving vehicle operating efficiency.

Description

Technical Field

[0001] This invention relates to the field of intelligent connected vehicle technology, specifically to a cloud-based vehicle speed planning system for new energy commercial vehicles based on artificial intelligence. Background Technology

[0002] Speed ​​planning directly impacts the driving comfort, economy, and transportation efficiency of commercial vehicles, and is primarily applied to PCC (Predictive Cruise Control) and PACC (Predictive Adaptive Cruise Control) systems. These systems acquire real-time map information (slope, curvature) from ADAS maps, and dynamically adjust the vehicle speed proactively with the goal of optimizing engine and transmission efficiency, avoiding unnecessary energy loss and achieving energy savings.

[0003] Publication number CN114148325A describes a predictive energy management method for heavy-duty hybrid commercial vehicles. It proposes a method that, after merging roads based on prior road gradients, utilizes a dynamic programming algorithm to integrate expected cruising speed and road speed limits to achieve predictively economically optimal vehicle speed planning for hybrid commercial vehicles. However, this scheme lacks big data technology and considers fewer factors, resulting in limited speed planning effectiveness on complex roads with many curves, such as mountain roads.

[0004] Publication number CN114523967A presents a neural network-based predictive cruise control method. This method uses road gradient and curve information from a high-precision map for the next time step, along with vehicle-to-vehicle speed data and current speed, as inputs to plan the vehicle speed. However, this scheme has poor applicability to new energy vehicles, and the use of fixed parameters for speed planning prevents continuous optimization of fuel economy.

[0005] With the development and application of 5G technology, big data analysis technology, and artificial intelligence technology in the commercial vehicle field, this invention proposes a cloud-based vehicle speed planning system for new energy commercial vehicles based on artificial intelligence algorithms. This system comprehensively extracts vehicle-side data and road features (slope, curve curvature, road grade, slope length), utilizes big data technology and random forest algorithm, integrates weather and road speed limits, and incorporates energy consumption parameters to achieve vehicle speed planning for new energy commercial vehicles that balances comfort, economy, and transportation efficiency, and continuously optimizes system performance based on the law of large numbers. Summary of the Invention

[0006] In view of the problems existing in the prior art, the purpose of this invention is to provide a cloud-based vehicle speed planning system for new energy commercial vehicles based on artificial intelligence.

[0007] The technical solution adopted by this invention to solve its technical problem is: a cloud-based vehicle speed planning system for new energy commercial vehicles based on artificial intelligence, including cloud data acquisition, collecting road-related data, and transmitting it to the data acquisition module;

[0008] Vehicle-side data collection: Collect vehicle-side route plans and transmit them to the data collection module. At the same time, upload the route plans to a multi-attribute map in the cloud.

[0009] The multi-attribute map generates GPS sequences along the road, and then finds the road attributes of the GPS point on the multi-attribute map based on the specified GPS, and transmits them to the data acquisition module.

[0010] The data acquisition module organizes and packages the collected cloud and vehicle data as input, and sends it to the algorithm model of the data processing module. The algorithm model analyzes the vehicle data and cloud data to plan the vehicle speed.

[0011] Preferably, the data processing module includes a road attribute confirmation unit, a road merging unit, a merged road structuring unit, and a planned vehicle speed generation unit. The road attribute confirmation unit determines the slope value S corresponding to each collected GPS point. i and curvature K i Roads are classified into 15 types based on their slope: gentle slope, gentle uphill slope, steep uphill slope, gentle downhill slope, and steep downhill slope. They are also classified into straight roads, gentle curves, and sharp curves based on their curvature. The distance L in front of the vehicle is divided into 15 road types.

[0012] Preferably, the road merging unit starts from the first set of GPS sequence data, merges roads with the same attributes, and the number of merged road GPS points does not exceed n. lim When encountering less than n lim The continuous GPS segments are also merged until the sequence is traversed to the last set of data. Finally, the length of each merged road segment is calculated using GPS points and added as a slope length attribute to the merged road.

[0013] Preferably, the merged road structure unit is such that after the roads are merged, the slope S of each merged road segment is such that... ai The average slope corresponding to all GPS points in the merged road segment, and the curvature K of each merged road segment. ai This represents the average curvature of the curves corresponding to all GPS points in the merged road segment, while also recording the road attributes of adjacent merged roads. None represents the starting or ending point of the planned road.

[0014] Preferably, after completing the structuring of merged roads, the planned vehicle speed generation unit combines vehicle-side data and inputs it into the algorithm model. After processing by the algorithm model, the vehicle speed of each merged road is obtained, and the last GPS point of each merged road is used as a control point to control the vehicle speed at that point as the planned vehicle speed.

[0015] Preferably, the algorithm model uses artificial intelligence algorithms to establish processing units by using n1% of the data stored in the database as the training set and n2% as the test set.

[0016] Preferably, the vehicle speed planning system also includes a database update module, which calculates the energy consumption of each merged road segment and identifies high-quality data. If, among samples with the same road characteristics in the database, the energy consumption of a sample is better than R%, it can be identified as high-quality data and included in the database for data updates.

[0017] Preferably, the database stores a number of structured, high-quality data, and every T0, the algorithm model is reconstructed using artificial intelligence algorithms and the stored data.

[0018] Preferably, the road-related data collected includes slope, road grade, curve curvature, number of lanes, and road speed limit.

[0019] The present invention has the following beneficial effects: After receiving a cloud-based vehicle speed planning request, the new energy commercial vehicle of the present invention can use the cloud-based vehicle speed planning system of the new energy commercial vehicle to plan the vehicle speed, calculate the vehicle speed that takes into account comfort, economy and transportation efficiency, reduce the driving intensity of users, reduce the energy consumption of vehicle driving, and improve vehicle operating efficiency. Attached Figure Description

[0020] Figure 1 This is an overall structural block diagram of the present invention.

[0021] Figure 2 This is a schematic diagram of the vehicle speed curve planned in the cloud for a certain section of road.

[0022] Figure 3 This is a schematic diagram illustrating the vehicle speed distribution and optimization of a sample library on a certain section of road. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0024] like Figure 1As shown, an artificial intelligence-based cloud-based vehicle speed planning system for new energy commercial vehicles comprises both hardware and software components. The hardware includes an algorithm controller, a CAN bus, and a T-BOX communication module, while the software includes a multi-attribute map and a cloud-stored database. The algorithm controller includes a data acquisition module and a data processing module.

[0025] The data acquisition module is responsible for collecting route planning, vehicle weight, continuous driving time, battery SOC, vehicle configuration information, system time, and weather data from the vehicle, and collecting road gradient, road grade, curve curvature, number of lanes, and road speed limits from the cloud. This data is then passed to the data processing module to generate the planned vehicle speed.

[0026] The main function of the data processing module is realized by the algorithm model generated by the cloud database through artificial intelligence algorithms. It can comprehensively analyze the received cloud and vehicle information to perform vehicle speed planning.

[0027] Multi-attribute maps are map data stored in the cloud that contain multiple road attributes (including but not limited to slope, road grade, curve curvature, number of lanes, and road speed limit). The road attributes under a given operating condition can be queried using a given GPS point or road ID.

[0028] The database stores a collection of high-quality data entries according to certain rules. Each entry contains various data obtained from the cloud and the vehicle itself. With the support of artificial intelligence algorithms, it can generate an algorithmic model for planning vehicle speed. Simultaneously, it accepts high-quality driving data to update its own database and subsequent algorithmic models.

[0029] High-quality data refers to driving data of vehicles with energy consumption better than n0% of other vehicles when driving on roads with the same vehicle type and road characteristics (where n0 is a calibration value with an initial value of 80).

[0030] The algorithm model utilizes artificial intelligence algorithms to establish a processing unit (where n1% is the training set and n2% is the test set, with n1 and n2 being calibrated values, n1 initially set to 70 and n2 initially set to 30). It receives information from the vehicle and cloud terminals collected by the data acquisition module, performs road merging and division according to certain rules, and outputs cloud-based vehicle speed planning based on the collected information to guide vehicle-side or provide feedback to the database for data updates.

[0031] Slope refers to the slope of a specified GPS point or the average slope of a path, measured using the percentage method.

[0032] Road classification refers to the functional level of a road, which is divided into expressways, Class I highways, Class II highways, Class III highways, and Class IV highways.

[0033] Curve curvature refers to the curvature of a road curve.

[0034] The number of lanes refers to the number of lanes for motor vehicles traveling in the same direction.

[0035] Road speed limits refer to the minimum / maximum speed limits for vehicles on a road.

[0036] Route planning involves storing the starting and ending points of the current trip, as well as the route information, on the vehicle's end.

[0037] Vehicle weight refers to the total weight of the vehicle and cargo calculated from the axle load at the vehicle end.

[0038] The continuous driving time is the time from the start of this trip, assuming no rest stops or rest periods of less than t0, to the present (t0 is a calibrated value, initially set to 20 minutes).

[0039] Battery SOC refers to the state of charge of a power battery.

[0040] Vehicle configuration data refers to a set of data that includes information such as vehicle model, main drive performance parameters, and design gross weight.

[0041] System time and weather refer to the local time and weather conditions during vehicle operation. The local time is accurate to the second, and the weather follows the system.

[0042] Example

[0043] A cloud-based vehicle speed planning system for new energy commercial vehicles completes the entire process of speed planning, including cloud data collection, vehicle-side data collection, road merging and division, algorithm model establishment, planned speed generation, and database and model updates. The specific process is as follows:

[0044] First, the cloud database stores a number of high-quality structured data. Every T0 (T0 is the standard quantity, with an initial value of 3 days), the algorithm model is reconstructed using artificial intelligence algorithms and stored data. (In the early stages of training the model, a smaller update cycle can be adopted to accelerate the iterative evolution of the database and the model. In the later stages, the update cycle can be appropriately slowed down to reduce the cloud load.)

[0045] Second, the cloud receives the cloud-based vehicle speed planning request from the relevant function, and begins to collect information such as the vehicle's route planning, vehicle weight, continuous driving time, battery SOC, vehicle configuration information, system time, and weather, and transmits it to the data acquisition module. Additionally, the route planning is uploaded to the cloud-based multi-attribute map.

[0046] Third, the multi-attribute map generates GPS sequences along the route based on the uploaded route planning, and then finds the road attributes of the GPS point on the multi-attribute map according to the specified GPS, including slope, road grade, curve curvature, number of lanes, and road speed limit, and transmits them to the data acquisition module.

[0047] Fourth, the data acquisition module organizes and collects data from the cloud and the vehicle, packages it, and sends it to the algorithm model as input to the data processing module.

[0048] Fifth, the data processing module processes the input in the following steps:

[0049] 1) Road Attribute Confirmation Unit

[0050] Based on the slope value Si and curve curvature Ki of each collected GPS point, roads can be classified into gentle slopes, gentle uphill slopes, steep uphill slopes, gentle downhill slopes, and steep downhill slopes according to slope; and into straight roads, gentle curves, and sharp curves according to curve curvature. The combination of these two classifications can divide roads with a distance L in front of the vehicle into 15 road types.

[0051] If the road slope corresponding to the GPS point is in [-S] G S G If the area is within the specified range, then that point is considered a level road.

[0052] If the road gradient is within [-S] v -S G If the point is within the range of a gentle downhill road, then that point is considered a gentle downhill road.

[0053] If the road slope is in (-∞, -S) v If the point is within the range of a steep downhill road, then that point is considered a steep downhill road.

[0054] If the road slope is (S) G S v If the area is within the specified range, then that point is considered a gently uphill road;

[0055] If the road slope is (S) v If the distance is within the range of ∞, then the point is considered as a steep uphill road;

[0056] If the curvature of the curve corresponding to the GPS point is in (0, K) G If the point is within the specified range, then that point is considered a straight path;

[0057] If the curvature of the curve is within (K) G K v If the point is within the specified range, then it is considered a gentle curve.

[0058] If the curvature of the curve is within (K) v If the value is within the range of (∞), then the point is considered a sharp bend;

[0059] Where L, S G S v K G K v For calibration purposes, the initial value of L is 1 km, and S... G The initial value is 3%, S v The initial value is 5%, KG The initial value is 0.001, K v The initial value is 0.02.

[0060] Based on the above rules, taking a specific section of road data as an example, the confirmation status of road attributes is shown in the table below:

[0061]

[0062]

[0063] 2) Road merging unit

[0064] To calculate slope length and reduce the number of road segments, roads with confirmed attributes are merged. Starting from the first set of GPS data, roads with the same attributes are merged, and the number of GPS points merged does not exceed n. lim n lim For calibration purposes, the initial value is 5). When encountering a value less than n... lim The continuous GPS segments were also merged until the last set of data was traversed. Finally, the length of each merged road segment was calculated using GPS points and added as the slope length attribute to the merged road. The specific results of merging the roads in 1) are as follows:

[0065]

[0066]

[0067] 3) Merge road structure units

[0068] After the road merging is completed, set the slope S of each merged road section. ai The average slope corresponding to all GPS points in the merged road segment, and the curvature K of each merged road segment. ai This represents the average curve curvature of all GPS points in the merged road segment, while also recording the road attributes of adjacent merged roads. "None" indicates the start or end point of the planned road. The specific results of structuring the merged roads based on the data in 2) are as follows:

[0069]

[0070]

[0071] 4) Planned vehicle speed generation

[0072] After structuring the merged road data, it is combined with vehicle-side data and input into the algorithm model. After processing by the algorithm model, a vehicle speed that balances comfort, economy, and transportation efficiency is obtained for each merged road. The last GPS point of each merged road is used as a control point, and the vehicle speed at that point is controlled as the planned speed. The specific results of the data in 3) after processing by the algorithm model are as follows:

[0073]

[0074] Next, the planned vehicle speed is calculated using the formula n. f Polynomial fitting of degree n f To calibrate the data, the initial value was 5, which yielded the vehicle speed curve generated in the cloud (see appendix). Figure 2 It can effectively adapt to needs such as rapid start, stable speed, hill climbing, cornering deceleration, and speed limiting. The obtained speed curve is output to the vehicle for intelligent drive adjustment.

[0075] Sixth, database update: After the data is sent to the vehicle and executed correctly, the energy consumption of each merged road segment is statistically analyzed to determine high-quality data. If, among samples with the same road characteristics in the database, the energy consumption is better than R% (R is a calibrated value, initially 80), it is considered high-quality data and added to the database for data updates. The vehicle speed distribution and optimization process of a certain road segment in the cloud database are shown in the attached figure. Figure 3 As shown. According to the law of large numbers, as high-quality samples accumulate, the speed distribution in the database will gradually approach the economically optimal speed on the same road segment, achieving the effect of gradual optimization of the database and algorithm model.

[0076] This invention is not limited to the above-described embodiments. Anyone should know that any structural changes made under the guidance of this invention, and any technical solutions that are the same as or similar to this invention, fall within the protection scope of this invention.

[0077] The technologies, shapes, and structures not described in detail in this invention are all known technologies.

Claims

1. A cloud-based vehicle speed planning system for new energy commercial vehicles based on artificial intelligence, characterized in that, This includes cloud-based data collection, which involves collecting road-related data and transmitting it to the data collection module. Vehicle-side data collection: Collect vehicle-side route plans and transmit them to the data collection module. At the same time, upload the route plans to a multi-attribute map in the cloud. The multi-attribute map generates GPS sequences along the road, and then finds the road attributes of the GPS point on the multi-attribute map based on the specified GPS, and transmits them to the data acquisition module. The data acquisition module organizes and packages the collected cloud and vehicle data as input, and sends it to the algorithm model of the data processing module. The algorithm model analyzes the vehicle data and cloud data to plan the vehicle speed.

2. The cloud-based vehicle speed planning system for new energy commercial vehicles based on artificial intelligence as described in claim 1, characterized in that, The data processing module includes a road attribute confirmation unit, a road merging unit, a merged road structuring unit, and a planned vehicle speed generation unit. The road attribute confirmation unit determines the slope value S corresponding to each collected GPS point. i and curvature K i Roads are classified into 15 types based on their slope: gentle slope, gentle uphill slope, steep uphill slope, gentle downhill slope, and steep downhill slope. They are also classified into straight roads, gentle curves, and sharp curves based on their curvature. The distance L in front of the vehicle is divided into 15 road types.

3. The cloud-based vehicle speed planning system for new energy commercial vehicles based on artificial intelligence according to claim 2, characterized in that, The road merging unit starts from the first set of GPS sequence data and merges roads with the same attributes, with the number of GPS points merged not exceeding n. lim When encountering less than n lim The continuous GPS segments are also merged until the sequence is traversed to the last set of data. Finally, the length of each merged road segment is calculated using GPS points and added as a slope length attribute to the merged road.

4. The cloud-based vehicle speed planning system for new energy commercial vehicles based on artificial intelligence according to claim 2, characterized in that, The merged road structure unit is the slope S of each merged road segment after the roads are merged. ai The average slope corresponding to all GPS points in the merged road segment, and the curvature K of each merged road segment. ai This represents the average curvature of the curves corresponding to all GPS points in the merged road segment, while also recording the road attributes of adjacent merged roads. None represents the starting or ending point of the planned road.

5. The cloud-based vehicle speed planning system for new energy commercial vehicles based on artificial intelligence according to claim 2, characterized in that, After structuring the merged roads, the planned vehicle speed generation unit combines vehicle-side data with the data input into the algorithm model. The algorithm model processes the data to obtain the vehicle speed for each merged road. The last GPS point of each merged road is used as a control point, and the vehicle speed at that point is controlled as the planned vehicle speed.

6. The cloud-based vehicle speed planning system for new energy commercial vehicles based on artificial intelligence according to claim 1, characterized in that, The algorithm model uses artificial intelligence algorithms to establish processing units by using n1% of the data stored in the database as the training set and n2% as the test set.

7. The cloud-based vehicle speed planning system for new energy commercial vehicles based on artificial intelligence according to claim 1, characterized in that, The vehicle speed planning system also includes a database update module, which calculates the energy consumption of each merged road segment and identifies high-quality data. If, among samples with the same road characteristics in the database, the energy consumption is better than R%, it can be identified as high-quality data and included in the database for data updates.

8. The cloud-based vehicle speed planning system for new energy commercial vehicles based on artificial intelligence according to claim 7, characterized in that, The database stores a number of structured, high-quality data sets. Every T0, the algorithm model is reconstructed using artificial intelligence algorithms and the stored data.

9. The cloud-based vehicle speed planning system for new energy commercial vehicles based on artificial intelligence according to claim 1, characterized in that, The road-related data collected includes slope, road grade, curve curvature, number of lanes, and road speed limit.

Images