WO2020029462A1

WO2020029462A1 - Self-driving system for electric vehicle

Info

Publication number: WO2020029462A1
Application number: PCT/CN2018/116004
Authority: WO
Inventors: 聂石启; 邱远红; 李春; 谭福伦; 何朝东; 司炎鑫; 黄苏杭; 陈苏荟; 顾治平
Original assignee: 金龙联合汽车工业（苏州）有限公司
Priority date: 2018-08-09
Filing date: 2018-11-16
Publication date: 2020-02-13
Also published as: CN108995538A

Abstract

A self-driving system for an electric vehicle, comprising: an environment sensing system which obtains, by processing data acquired by various sensors, position information about an obstacle, road image information, obstacle image information, and information about a position and speed relative to an obstacle; a positioning and navigation system which uses a gyroscope in a combined positioning and navigation system to determine a vehicle attitude, so as to perform self-positioning, and constructs an incremental map on the basis of self-positioning, so as to travel along a pre-determined route; a motion control system which longitudinally controls a vehicle by means of the cooperation of an accelerator and a brake, so as to control a vehicle speed, transversely controls a vehicle by means of a steering system, so as to control a path of a vehicle, and recovers energy by means of a control strategy when performing deceleration control on a vehicle; and a driving controller which generates a control strategy according to the acquired data and vehicle parameters, so as to control travelling of a vehicle. The present invention can recover energy to the maximum extent, improving the economy and dynamic performance of an electric vehicle.

Description

Unmanned driving system for electric vehicle

Technical field

The invention belongs to the technical field of unmanned driving, and particularly relates to an unmanned driving system of an electric vehicle.

Background technique

The current autonomous driving technology is still in the experimental stage. The reason why it cannot enter the practical stage as soon as possible is because the driving environment is too complicated, and the current level of artificial intelligence cannot reach the level that humans perceive complex environments. Take Google ’s self-driving car as an example. Its main key components include GPS chips, cameras, radar, high-performance processors, and so on. Among them, GPS is used to provide accurate map information, which is currently obtained by manual driving; cameras and radar are equivalent to human eyes, and are used to sense the surrounding environment of the car, including identifying lane lines, signal lights, and surrounding obstacles; for high-performance processors It is used to process the information collected by GPS, camera, radar, etc. and issue instructions to the throttle, brake, steering wheel and other execution components. This kind of automatic driving system is costly and requires strong and accurate map data. At present, there are still inability to drive on snow, maps cannot include buildings and road modifications in a timely manner, failure to make judgments when encountering construction and accidents, and failure to identify traffic police. Gestures and language are difficult. Machines are good at simple, deterministic, and repetitive tasks. They are not good at dealing with complex, uncertain, and changeable tasks. In a short period of time, machines cannot reach the perception and judgment capabilities of human beings, and cannot independently respond to all special situations encountered during driving.

At present, intelligent driving is the development direction of electric vehicles. Intelligent driving enables the complexity of electric vehicle driving tasks to be effectively controlled, which helps the driver to correct bad operating habits and correct wrong operations, and to give full play to the economics and power of electric vehicles.

Summary of the invention

In view of the technical problems mentioned above, an object of the present invention is to provide an unmanned driving system for an electric vehicle. On the premise of ensuring safety and conforming to the driver's braking habits, recovering as much energy as possible can improve the electric vehicle. Economical and dynamic.

The technical solution of the present invention is:

An unmanned driving system for an electric vehicle includes:

An environment perception system that processes the data collected by various sensors to obtain obstacle position information, road image information, obstacle image information, and relative position and speed information with obstacles;

A positioning and navigation system, which uses the gyroscope in the integrated positioning and navigation system to determine the attitude of the vehicle, and tracks and navigates according to the vehicle's time and position, attitude, driving direction and speed information, or according to a predetermined storage path planning data, or during the movement process Self-positioning based on location estimates and maps, while building incremental maps on the basis of self-positioning, and driving along predetermined routes;

A motion control system, which controls the vehicle longitudinally through the cooperation of the accelerator and brakes, controls the vehicle speed, controls the vehicle's path laterally through the steering system, and controls the vehicle's path; during vehicle deceleration control, energy is recovered through control strategies;

A driving controller, based on the collected data and vehicle parameters, develops a control strategy to control the driving of the vehicle.

In a preferred technical solution, the sensors of the environment sensing system include 4 laser radars, 1 millimeter wave radar, and 2 cameras.

In a preferred technical solution, the control strategy of energy recovery includes: when a moderate speed reduction or a long downhill driving is required, the driving controller controls the motor controller to perform the torque-limiting driving of the motor, and the motor braking is used as the main braking; When the braking torque of the motor is less than the braking force of the driving controller, the driving controller's control wire control valve is adjusted as the main brake and the motor brake is used as the auxiliary brake; a parallel energy recovery method is used to generate electrical power in the entire motor. During the dynamic process, the drive motor runs in the power generation state, and the driving controller communicates with the battery management system to monitor and control the charging current and charging time during the charging process.

In a preferred technical solution, the optimal coverage interval of mechanical braking and motor braking is analyzed through data, and motor braking is used in the coverage interval for energy recovery.

In a preferred technical solution, the driving controller determines a driving mode, and the driving mode includes a manual driving mode and an unmanned driving mode.

Compared with the prior art, the beneficial effects of the present invention are:

1. On the premise of ensuring safety and conforming to the driver's braking habits, recovering as much energy as possible can improve the economics and power of electric vehicles.

2. The positioning and navigation system builds an incremental map on the basis of its own positioning to realize unmanned driving and navigation, which can be unmanned according to a predetermined route or unmanned in an unknown environment.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention is further described below with reference to the drawings and embodiments:

FIG. 1 is a structural diagram of an unmanned driving system of an electric vehicle according to the present invention;

FIG. 2 is a logic diagram of a dual-mode driving work switch according to the present invention.

detailed description

In order to make the objectives, technical solutions, and advantages of the present invention clearer, the following further describes the present invention in detail with reference to specific embodiments and the accompanying drawings. It should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the invention. In addition, in the following description, descriptions of well-known structures and techniques are omitted to avoid unnecessarily obscuring the concept of the present invention.

Example:

As shown in Figure 1, an unmanned driving system for an electric vehicle includes:

1.Environmental awareness system

The environment sensing system uses various sensors to collect data on the environment, obtain driving environment information, and process the data in the information. The environment sensing system provides the location information of the surrounding obstacles for the electric vehicle, as well as the relative distance and relative speed with the surrounding vehicles, road settings or obstacles such as pedestrians, and then provides information basis for control decisions. The invention integrates the working principles, detection ranges, advantages and disadvantages of the working conditions of each sensor, etc., and adopts a redundant design scheme. The invention adopts four 16-line laser radars, one 77G long / short-range millimeter-wave radar, and two on-board vision cameras as an environment sensing system for driverless passenger cars. Among them, four 16-wire lidars use Ethernet to communicate with the unmanned controller. One 77G long / short-range millimeter-wave radar uses a high-speed CAN bus with a baud rate of 500 kbit / s to communicate with the unmanned controller. Two The vehicle vision camera communicates with the driverless controller using a low-speed CAN bus expansion frame with a baud rate of 250 kbit / s.

2.Positioning and navigation system

The positioning and navigation system is used to provide information such as the position and attitude of the vehicle. The present invention installs two DGPSs with a regional augmentation system to improve positioning accuracy, uses a gyroscope in a combined positioning and navigation system to determine vehicle attitude, uses tracking estimation technology and SLAM technology, according to the momentary position, attitude, and driving of an unmanned vehicle Direction and speed information, or plan and track data according to a predetermined storage path to navigate or drive, or locate itself based on position estimates and maps during the movement, and build an incremental map based on its own positioning to achieve unmanned driving and navigation , You can drive autonomously in accordance with a predetermined route or in an unknown environment.

3. Motion control system

Vehicle motion control is divided into longitudinal control and lateral control. Longitudinal control is the precise control of vehicle speed through the coordination of throttle and brake. Lateral control is the control of the vehicle's path through steering control. The invention adopts a linear control throttle for acceleration control; adopts a linear control air valve for deceleration air pressure braking, and comprehensively controls the electronic hand brake, the braking ability of the motor and the energy recovery ability control strategy, and the specific control strategy includes: turning on unmanned driving in the vehicle In the mode, the unmanned controller controls the release of the electronic handbrake, and controls the wire-controlled throttle to accelerate or run smoothly. When a moderate or slow down or long downhill driving is required, the unmanned controller instructs the motor controller to perform the motor limit. Torque driving, the motor brake is used as the main brake; when the motor's braking torque does not meet the braking force of the unmanned controller, the unmanned controller commands the line control valve to be adjusted as the main brake. The electric mechanism Action is auxiliary braking. The invention adopts a parallel energy recovery method. During the entire process of electric brake generation of the motor, the driving motor runs in a power generation state, and a portion of the kinetic energy is fed back to the battery to charge it. The driverless controller and battery management system (BMS) During data communication, the charging current or charging time during the charging process is monitored and controlled to avoid factors such as excessive charging current or charging time during energy recovery from affecting the battery life. In the entire control strategy, always consider braking safety as the control strategy with the highest priority. Through data analysis and algorithm control, find the best coverage area of mechanical braking and motor braking to ensure safety and meet the driver's braking. Under the premise of habit, recover as much energy as possible.

Acceleration and deceleration of the vehicle can be accurately controlled longitudinally through CAN bus communication. The invention adopts an electro-hydraulic linear control steering system, and combines the steering angle sensor and the torque sensor on the steering gear to obtain steering information and gyroscope judgment information on the vehicle attitude. The CAN bus is used to realize the lateral control of the vehicle and ensure the tracking of the vehicle. Path control.

4.Unmanned controller

The unmanned controller adopts a modular design concept, integrating environmental awareness module, positioning and navigation module, planning system module, vehicle communication module and control system module, etc., and the modules use Ethernet and CAN bus communication.

The unmanned controller uses a reconfigurable computing AI chip, which is programmed based on the Linux system platform, and customizes the control strategy based on the parameters of the entire bus and operating characteristics. Specific control strategies include:

(1) The unmanned controller serves as the arbitration controller for the control mode, and determines whether it is currently in the unmanned mode and whether it meets the conditions of the unmanned mode.

(2) When in the manual driving mode, the control system module shields the control command issued by the unmanned controller, and responds to the manual operation command.

(3) When in the unmanned control mode, according to the information of the lidar, millimeter-wave radar and camera received by the unmanned controller received by the environmental sensing module, the optimized information fused by each sensor, and The body attitude information sensed by the gyroscope and the latitude and longitude information provided by the DGPS positioning and navigation module are combined with the tracking route learned through deep learning, or the adaptive route planning travel is performed through parallel mapping and positioning technology (SLAM).

(4) The control system module responds to the execution request instructions of the electronic handbrake, gear position, remote control throttle, remote control and remote control, etc. issued by the unmanned controller.

(5) Trackless driving on closed roads can be achieved in the unmanned mode.

(6) In unmanned mode, open road driving can be realized. When encountering a static obstacle in front, you can drive on the left side of the road by yourself. If you detect an obstacle on the left lane or the lane is too narrow, you can automatically stop and wait.

(7) In unmanned mode, open road driving can be realized. When encountering a dynamic obstacle ahead, drive on the left side of the road. If it is detected that there is an obstacle on the left lane or the lane is too narrow, the vehicle can follow the front obstacle by itself.

(8) In the unmanned mode, when a traffic light at an intersection is detected, it can stop, wait, start, run, etc. according to the status of the traffic light.

(9) Through the vehicle communication module, the vehicle's CAN bus is issued to the vehicle's CAN bus to execute commands such as headlights, turn signals, brake lights, reversing lights, door switches, horn switches and other vehicle driving status components.

5.Dual driving mode

The present invention adopts an unmanned driving mode and a normal driving mode to switch between dual driving modes. The dual driving mode switching strategy is as follows. FIG. 2 is a logic diagram of dual mode driving work switching.

When it is necessary to enter the "unmanned mode" from the "normal driving mode", press the external unmanned control switch to send a request 2 to the unmanned controller. If the unmanned controller judges whether it can enter unmanned at this time Driving mode, allowed to enter unmanned mode, execute 3, not allowed to enter unmanned mode, feedback signal 4;

When you need to enter the “normal driving mode” from the “unmanned mode”, you can switch to the normal driving mode by resetting the external unmanned control switch or the pre-planned exit driver logic, and execute the control instruction 1.

It should be understood that, the foregoing specific implementation manners of the present invention are only used to exemplarily illustrate or explain the principle of the present invention, and do not constitute a limitation to the present invention. Therefore, any modification, equivalent replacement, improvement, etc. made without departing from the spirit and scope of the present invention should be included in the protection scope of the present invention. Further, the appended claims of the present invention are intended to cover all changes and modifications that fall within the scope and boundary of the appended claims, or equivalents of such ranges and boundaries.

Claims

An unmanned driving system for an electric vehicle, comprising:

An environment perception system that processes the data collected by various sensors to obtain obstacle position information, road image information, obstacle image information, and relative position and speed information with obstacles;

A positioning and navigation system, which uses the gyroscope in the integrated positioning and navigation system to determine the attitude of the vehicle, and tracks and navigates according to the vehicle's time and position, attitude, driving direction and speed information, or according to a predetermined storage path planning data, or during the movement process Self-positioning based on location estimates and maps, while building incremental maps on the basis of self-positioning, and driving along predetermined routes;

A motion control system, which controls the vehicle longitudinally through the cooperation of the accelerator and brake, controls the vehicle speed, controls the vehicle's path laterally through the steering system, and controls the vehicle's path; during vehicle deceleration control, energy recovery is achieved through control strategies;

A driving controller, based on the collected data and vehicle parameters, develops a control strategy to control the driving of the vehicle.
The unmanned driving system for an electric vehicle according to claim 1, wherein the sensors of the environment perception system include 4 lidars, 1 millimeter wave radar, and 2 cameras.
The unmanned driving system for an electric vehicle according to claim 1, wherein the control strategy for energy recovery comprises: when a moderate speed reduction or a long downhill driving is required, the driving controller controls the motor controller to execute the motor limit Twisting driving, the motor brake is used as the main brake; when the braking torque of the motor is less than the braking force of the driving controller, the driving controller control wire valve is adjusted as the main brake, and the motor brake is used as the auxiliary brake; The parallel energy recovery method is used. During the entire process of electric braking of the motor, the driving motor runs in the power generation state. The driving controller and the battery management system perform data communication to monitor and control the charging current and charging time during charging. .
The unmanned driving system for an electric vehicle according to claim 3, wherein the optimal coverage interval of mechanical braking and motor braking is analyzed through data, and motor braking is used in the coverage interval to perform energy recovery.
The unmanned driving system for an electric vehicle according to claim 1, wherein the driving controller determines a driving mode, and the driving mode includes a manual driving mode and an unmanned driving mode.