CN-121989946-A - New energy self-adaptive driving mode switching system based on multi-sensor fusion

CN121989946ACN 121989946 ACN121989946 ACN 121989946ACN-121989946-A

Abstract

The invention relates to the technical field of intelligent control of vehicles, and particularly discloses a new energy self-adaptive driving mode switching system based on multi-sensor fusion, wherein an AEB system integrated camera is used for acquiring 50-100m pre-aiming road surface image information in front of a vehicle; the scene recognition module is connected with the data acquisition module, the scene recognition module recognizes the current load state, gradient working condition, real-time road surface grade and pre-aiming road surface grade based on the acquired data, the driving mode decision module stores preset 5 driving modes and corresponding power parameters and cab environment parameters, a target driving mode is output according to the principle of pre-aiming priority and safety priority, the execution feedback module comprises a power control unit, a brake control unit, a cab environment control unit and an instrument display unit, and the execution feedback module is adaptive to a light truck urban distribution scene, reduces cost and increases efficiency, is safe to upgrade, prevents pre-aiming, optimizes cab experience, meets the requirement of a driver, has zero additional cost and is high in reliability.

Inventors

WANG YONG
FANG HUIPING
WANG XIAOLONG
HUANG XINSHUO
YUAN XIAOQI
LI BO
TANG QIAN

Assignees

陕汽集团商用车有限公司
陕西汽车集团股份有限公司

Dates

Publication Date: 20260508
Application Date: 20260209

Claims (10)

1. A new energy self-adaptive driving mode switching system based on multi-sensor fusion is characterized by comprising a data acquisition module, a scene recognition module, a driving mode decision module and an execution feedback module, The data acquisition module comprises an EPB controller, a VCU controller, an MCU controller, an ABS controller and an AEB system, wherein an AEB system integrated camera is used for acquiring 50-100m pre-aiming road surface image information in front of a vehicle, and each controller synchronously uploads gradient angle, motor torque, battery power, wheel slip rate and road surface image data through a CAN bus; The scene recognition module is connected with the data acquisition module, and is used for recognizing the current load state, gradient working conditions, real-time road surface grades and pre-aiming road surface grades based on the acquired data, wherein the load state comprises light load, medium load and heavy load, the gradient working conditions comprise flat, gentle slope and abrupt slope, and the pre-aiming road surface grades are judged based on the AEB system image characteristic recognition result; The driving mode decision module stores 5 preset driving modes and corresponding power parameters and cab environment parameters, wherein the driving modes comprise an economic mode, a standard mode, a power mode, a safety mode and a comfort mode, and the driving mode decision module outputs a target driving mode according to the principle of pretightening priority and safety priority based on the current scene and pretightening scene output by the scene recognition module; The execution feedback module comprises a power control unit, a brake control unit, a cab environment control unit and an instrument display unit and is used for executing a mode switching instruction and feeding back the current mode state and the switching reason to a driver.
2. The new energy source adaptive driving mode switching system based on multi-sensor fusion according to claim 1, wherein the data acquisition module: the AEB system identifies the type of the pre-aiming road surface through image features, and comprises the steps of judging a wet road surface through ponding reflection features, judging a non-paved road surface through broken stone texture features and judging an ice and snow road surface through white coverage features; The ABS controller judges the grade of the real-time road surface through the wheel slip rate, wherein the slip rate is more than 20 percent of the wet road surface, and the slip rate is 5 to 10 percent of the good road surface; the MCU controller and the VCU controller cooperatively judge the load state, namely, under the constant-speed working condition, the motor torque is greater than rated torque a and the battery power is greater than rated power b and is heavy load, the motor torque is less than rated torque a1 and the battery power is less than rated power b1 and is light load, and the balance is medium load, wherein a=60%, b=50%, a1=30% and b1=30%.
3. The new energy source adaptive driving mode switching system based on multi-sensor fusion according to claim 2, wherein the preset mode of the driving mode decision module satisfies: The motor output torque is limited at rated value d, d=80%, the ABS value is reduced by 20%, the braking pressure is increased by 15%, the vibration damping of the cab seat is adjusted to be high, and the instrument is backlit; The comfortable mode comprises that motor output torque fluctuation is less than or equal to 5%, energy recovery intensity is less than or equal to 10%, ventilation and heating of a cab seat are automatically started based on ambient temperature, ventilation is started at a temperature of 28 ℃, heating is started at a temperature of 15 ℃, steering assistance of a steering wheel is adjusted to be light, and when a user approaches to the front for 1-3s, the user enters a wet road or a non-paved road or an ice-snow road, safety mode switching preparation is triggered by 0.5-1s in advance.
4. The new energy self-adaptive driving mode switching system based on multi-sensor fusion according to claim 1, wherein the instrument display unit comprises a pre-aiming road surface image thumbnail, and the switching reasons are synchronously displayed during mode switching.
5. The new energy self-adaptive driving mode switching system based on multi-sensor fusion according to claim 1, wherein the adjustment parameters of the cab environment control unit comprise air conditioning air quantity, seat ventilation or heating state, seat vibration damping level and steering wheel steering power assisting level, the adjustment response time is less than or equal to 2s, and the air conditioning air quantity is 1-3 gears.
6. The new energy source adaptive driving mode switching method based on the multi-sensor fusion is realized by adopting the new energy source adaptive driving mode switching system based on the multi-sensor fusion as claimed in any one of claims 1 to 5, and is characterized by comprising the following steps: s1, data acquisition and preprocessing, wherein gradient, load, real-time road surface and pre-aiming road surface data are synchronously acquired through an EPB controller, a VCU controller, an MCU controller, an ABS controller and an AEB system, fluctuation is eliminated through sliding window filtering, and the size of a sliding window is 3S; s2, scene recognition, namely, recognizing a pre-aiming road surface grade through AEB system image characteristics, judging a loading state, a gradient working condition and a real-time road surface grade by combining real-time data, and outputting a composite scene every 0.5S; S3, mode decision, namely, if the pre-aiming road surface is a complex road condition, a wet road surface or a non-paved road surface or an ice and snow road surface is prepared to be switched to a safe mode in advance for 0.5-1S, and if the pre-aiming road surface is good, an economic mode or a standard mode or a power mode or a comfortable mode is matched according to a load and gradient; s4, executing and feeding back, finishing power parameter or braking parameter adjustment within 100ms, finishing cab environment parameter adjustment within 2S, and synchronously displaying mode state, switching reasons and pre-aiming road surface prompt by the instrument.
7. The method for switching new energy source adaptive driving modes based on multi-sensor fusion according to claim 6, wherein the validity check rule of the data in the step S1 is that when the gradient angle is >30 degrees or the AEB system image has no valid road surface feature, the average value of the valid data of the previous 3 times is adopted to replace the abnormal data.
8. The method for switching new energy sources to adaptive driving modes based on multi-sensor fusion according to claim 6, wherein the logic decision rule for scene recognition in step S2 comprises: Slope working condition, namely slope angle < c flat road surface, slope angle c 1 is a gentle slope, slope angle > c 2 is a steep slope, wherein c=3 degrees, c 1 =3°~15°,c 2 =15 degrees, the combination of vehicle speed drop is more than 5km/h, and torque rise is judged to be a climbing working condition; the road surface grade priority is that the pre-aiming road surface grade is higher than the real-time road surface grade, and when the pre-aiming reaches the complex road condition, the mode switching is triggered preferentially based on the pre-aiming result.
9. The new energy source adaptive driving mode switching method based on multi-sensor fusion according to claim 6, wherein the mode matching rule of the mode decision in step S3 comprises: light load + flat good paved road surface → economic mode; Medium load and flat or gentle slope well pave the road surface and the driving time is longer than 1h- & gt comfort mode; heavy load + steep/climbing road surface → power mode; real-time or pre-aiming wet road or non-paved road or ice and snow road to a safe mode.
10. The method for switching the new energy source adaptive driving modes based on the multi-sensor fusion according to claim 6, wherein the mode rechecking mechanism for the mode decision in the step S3 is to recheck the pre-aiming data and the real-time data once every 5S when the scene is unchanged, and control the error switching caused by the instantaneous fluctuation of the sensor.

Description

New energy self-adaptive driving mode switching system based on multi-sensor fusion Technical Field The invention belongs to the technical field of intelligent control of vehicles, and particularly relates to a new energy self-adaptive driving mode switching system based on multi-sensor fusion. Background The new energy light truck is used as a main force of urban transportation, and the daily operation needs to cope with complex and changeable scenes that heavy load starting is achieved in a logistics park, medium load uniform speed running is achieved in urban roads, suburban ramp sections are frequently traversed, and complex road conditions such as wet and slippery rainy days, non-paved roads in construction sections and the like are often encountered. At present, the switching of the driving modes of the new energy light truck is manually operated by a driver, and two main core pain points exist: The mode is mismatched with the scene, namely the motor is easy to be frequently overloaded and suddenly lowered due to no switching to the power mode when the motor climbs a slope under heavy load, and the energy consumption is increased by 15-20% by using the heavy load mode when the motor is in light load return, so that the urban distribution cost and efficiency requirements are violated. The operation load of a driver is heavy, namely the driver is required to take loading and unloading, route planning and mode switching into consideration when the driver starts and stops for more than 50 times on a city light truck day, and potential safety hazards such as insufficient power (flameout on an ascending slope) or longer braking distance (heavy-load uncut safety mode) are caused due to easy inadvertent wrong mode selection. In the prior art, the partial driving mode switching scheme relies on additional sensors, so that the cost of a bicycle is increased by 500-1200 yuan, and the failure rate of the sensors exceeds 15% due to limited space of a light truck chassis and frequent jolt vibration. Along with the upgrade of new energy light-truck electronic architecture, EPB, VCU, MCU, ABS and AEB systems realize CAN bus interconnection, and how to utilize the existing hardware redundancy data and combine the pavement pre-aiming capability of the AEB camera to construct a data-scene-mode automatic matching logic becomes a key for solving the problems. The system and the method for switching the self-adaptive driving mode of the new energy light truck, which are based on the cooperative work of an EPB (electronic parking controller), a VCU (vehicle control unit), an MCU (motor controller), an ABS (anti-lock braking system) and an AEB (automatic emergency braking) system, are provided. Disclosure of Invention The invention aims to provide a new energy self-adaptive driving mode switching system based on multi-sensor fusion, which is free from additional hardware, and is used for constructing a scene logic model by fusing EPB, VCU, MCU, ABS data and AEB system road surface image information, automatically identifying the load state, road gradient and road surface grade (including real-time and pre-aiming road conditions) of a new energy light card, outputting an optimal driving mode and improving driving safety and operation economy. In order to solve the problems in the prior art, the technical scheme adopted by the invention is that the new energy source adaptive driving mode switching system based on multi-sensor fusion comprises a data acquisition module, a scene recognition module, a driving mode decision module and an execution feedback module, The data acquisition module comprises an EPB controller, a VCU controller, an MCU controller, an ABS controller and an AEB system, wherein an AEB system integrated camera is used for acquiring 50-100m pre-aiming road surface image information in front of a vehicle, and each controller synchronously uploads gradient angle, motor torque, battery power, wheel slip rate and road surface image data through a CAN bus; The scene recognition module is connected with the data acquisition module, and is used for recognizing the current load state, gradient working conditions, real-time road surface grades and pre-aiming road surface grades based on the acquired data, wherein the load state comprises light load, medium load and heavy load, the gradient working conditions comprise flat, gentle slope and abrupt slope, and the pre-aiming road surface grades are judged based on the AEB system image characteristic recognition result; The driving mode decision module stores 5 preset driving modes and corresponding power parameters and cab environment parameters, wherein the driving modes comprise an economic mode, a standard mode, a power mode, a safety mode and a comfort mode, and the driving mode decision module outputs a target driving mode according to the principle of pretightening priority and safety priority based on the current scene and pretightening scene output by the scene re