CN107031600A - Automated driving system based on highway - Google Patents

Abstract

The invention relates to the technical field of automatic driving, in particular to a highway-based automatic driving system. Including sensors, positioning modules, human-computer interaction modules, automatic driving controllers, and vehicle execution modules. Monocular camera, stereo camera, forward long-range millimeter microwave radar and four-corner medium-range millimeter microwave radar are used to replace the traditional laser radar, and are respectively arranged behind the interior rearview mirror, in the center of the vehicle's front bumper, and on both sides of the front and rear bumpers. Location, separate collection of various information in traffic sign road information and obstacle information, and integrate a complete set of automatic driving systems such as perception fusion, path planning, decision planning and control into an embedded controller. Compared with the traditional automatic driving system using lidar, it is low in cost and easy to integrate. Its perception accuracy is greatly improved, the system integration is high, and the size is small, which greatly reduces the size and power consumption of the automatic driving system, and is easy to mass produce.

Description

Highway-based autonomous driving system

technical field

The invention relates to the technical field of automatic driving, in particular to a highway-based automatic driving system.

Background technique

Benefiting from the rapid progress of processor chips, deep learning technology and sensor technology, autonomous driving research has developed rapidly. Many OEMs, component suppliers, Internet companies and scientific research institutes have successively launched autonomous driving prototypes or autonomous driving systems, such as Google's driverless prototype, Baidu's Baizhi autonomous driving prototype, Nissan's drivepilot 1.0 automatic driving system, etc. The automatic driving system in the related technology has the following problems: the sensor uses lidar to perceive the driving environment, but at this stage the lidar is expensive, and it is not easy to integrate into the car body, affecting the shape; Integration of control and other systems into one controller requires multiple industrial computers or multiple controllers to jointly complete the automatic driving control function. The system has a large size, high power consumption, and anti-vibration performance that does not meet the requirements of vehicle regulations. It is not easy to commercialize and quantify Produce.

Contents of the invention

In order to solve the above technical problems, the present invention provides a highway-based embedded integrated automatic driving controller with high integration, low cost and easy quantitative production.

The technical scheme of the present invention is, comprises:

Sensor: used to obtain traffic sign road information and obstacle information, and send it to the automatic driving controller;

Positioning module: used to calculate the absolute position information of the own vehicle and the high-precision map information near the absolute position of the car wash according to GPS inertial navigation and high-precision map information, and send them to the automatic driving controller;

Human-computer interaction module: used to receive driver instructions, and display vehicle status information, road environment perception information, and information related to the automatic driving system during automatic driving;

Automatic driving controller: used to receive the traffic sign road information and obstacle information sent by the sensor, and fuse it with the high-precision map information sent by the positioning module to establish a local environment map for the vehicle to drive. Starting point and end point, plan a local driving route that meets traffic rules and safety requirements on the local vehicle driving environment map established by the multi-sensor information fusion module, and finally send control instructions to the vehicle execution module according to the local driving route;

Vehicle execution module: used to receive the control instruction sent by the automatic driving controller, and execute corresponding actions.

Further, the sensor includes a monocular camera, and the monocular camera is used to receive traffic sign road information, and the traffic sign road information includes at least traffic sign information, traffic signal light information, lane line information, road edge information, road guidance information.

Further, the sensor includes a stereo camera, a forward long-range millimeter microwave radar and a four-corner middle-range millimeter microwave radar, and the stereo camera, the forward long-range millimeter microwave radar and the four-corner middle-range millimeter microwave radar are used for Obstacle information is received, and the obstacle information includes at least pedestrian information, vehicle information, and other obstacle information.

Further, the monocular camera and the stereo camera are installed behind the interior rearview mirror.

Further, the forward long-range millimeter-wave radar is installed in the center of the front bumper of the vehicle, and the four-corner mid-range millimeter-wave radar includes the front left-corner mid-range millimeter-wave radar, the front right-corner mid-range millimeter-wave radar, and the rear left corner mid-range millimeter-wave radar. The millimeter-wave radar and the middle-range millimeter-wave radar in the rear right corner are respectively installed on both sides of the front and rear bumpers.

Further, the automatic driving controller includes an automatic driving perception unit integrated in an editable gate array, a vehicle vertical and lateral movement control module, a driving mode switching control module and a human-computer interaction control module, and the editable gate array includes A data transmission interface capable of at least data interaction with sensors and positioning modules, a control signal transmission interface capable of sending control commands to the vehicle executive module, and a human-computer interaction interface capable of signal transmission with the human-computer interaction module;

The automatic driving perception unit is used to collect traffic sign road information and obstacle information through a data transmission interface, and to fuse with high-precision maps, GPS inertial navigation information, wheels, and vehicle speed information to generate local driving path information, and to generate local driving path information. The path information is sent to the vehicle longitudinal and lateral movement control module, and the obstacle position information and the vehicle position information are sent to the human-machine interaction control module;

The vehicle longitudinal and lateral motion control module is used to receive the vehicle's own attitude information sent by the local driving path and the human-computer interaction control module, and send control instructions to each module of the execution unit through the control signal transmission interface;

The driving mode switching control module is used to monitor the input of the human-computer interaction control module, control the switching between the automatic driving mode and the manual driving mode, and send the switched vehicle's own attitude information to the vehicle vertical and horizontal motion control module;

The human-computer interaction control module is used to receive the control instructions sent by the human-computer interaction module through the human-computer interaction interface, as well as the obstacle position information and the self-vehicle position information sent by the automatic driving perception unit, and send them to people through the human-computer interaction interface. The computer interaction module displays the output.

Further, the automatic driving perception unit includes a global path planning module, the global path planning module is used to receive the start point and end point set by the driver in the human-computer interaction module, and complete the process according to the vehicle position information sent by the vehicle fusion positioning module. Global path planning for autonomous driving behavior including driving routes.

Further, the automatic driving perception unit includes a local path planning module, the local path planning module receives the driving route planned by the global path planning module and the vehicle position information sent by the vehicle fusion positioning module, and establishes On the local environment map where the vehicle travels, plan a local driving route that meets traffic rules and safety requirements, and send the local driving route to the vehicle longitudinal and lateral movement control module.

Further, the automatic driving controller also includes an exception processing module, the abnormal processing module is used to monitor the vehicle state, the automatic driving controller and road weather conditions, accept the alarm signal for processing, and send the alarm signal to people Computer interactive control module.

Further, the vehicle execution module includes vehicle EMS, EPS, ESC, EPB, BCM and TCU, which are used to control the throttle opening, steering, braking, parking, gear position, lights and horn according to the control instructions sent by the automatic driving controller. Take control.

Beneficial effects of the present invention: use monocular camera, stereo camera, forward long-distance millimeter microwave radar and four-corner mid-distance millimeter microwave radar to replace traditional laser radar, and arrange them respectively behind the interior rearview mirror and in the center of the front bumper of the vehicle , front and rear bumpers, separate collection of various information in traffic signs, road information and obstacle information, and integrate a complete set of automatic driving systems such as perception fusion, path planning, decision planning and control into an embedded controller accomplish. Compared with the traditional automatic driving system using lidar, it is low in cost and easy to integrate. Its perception accuracy is greatly improved, the system is highly integrated, and its size is small. It can realize automatic driving functions such as adaptive cruise, lane keeping, autonomous lane change, autonomous overtaking, autonomous acceleration, emergency braking, and automatic curb parking. The size and power consumption of the autonomous driving system are reduced, and it is easy to mass-produce and has higher shock resistance. Compared with the data transmission between traditional equipment and equipment, its transmission rate is faster and the calculation accuracy is higher. Due to fewer external interfaces, the failure rate is low, and the control stability of the system is higher.

Description of drawings

Fig. 1 is a module connection diagram of the present invention;

Fig. 2 is a connection diagram of the automatic driving controller module of the present invention.

detailed description

Below by specific embodiment the present invention is described in further detail:

As shown in Figure 1, the present invention consists of five parts: a sensor, a positioning module, a human-computer interaction module, a vehicle execution module and an automatic driving controller.

The sensor consists of low-cost sensors such as vision sensors and millimeter-wave radars. Vision sensors use monocular cameras and stereo cameras, while millimeter-wave radars include forward long-range millimeter-wave radars and four-corner mid-range millimeter-wave radars. Four-corner mid-range millimeter-wave radars Including the middle-range millimeter-wave radar in the front left corner, the middle-range millimeter-wave radar in the front right corner, the middle-range millimeter-wave radar in the rear left corner, and the middle-range millimeter-wave radar in the rear right corner. Among them, the monocular camera and the stereo camera are installed behind the interior rearview mirror, the forward long-range millimeter-wave radar is installed in the center of the front bumper of the vehicle, the middle-range millimeter-wave radar in the front left corner, the middle-range millimeter-wave radar in the front right corner, and the rear left The corner-middle-range millimeter-wave radar and the rear-right corner-middle-range millimeter-wave radar are respectively installed at the four corners of the vehicle (that is, both sides of the front and rear bumpers). The vertical measurable distance of the monocular camera should be more than 100 meters, the vertical measurable distance of the stereo camera should be more than 40 meters, and the horizontal viewing angle should be more than 50 degrees; More than 20 degrees, the longitudinal measurable distance of the four-corner mid-range millimeter-wave radar is more than 50 meters, and the horizontal viewing angle is more than 120 degrees. The monocular camera is used to receive traffic sign road information, and the traffic sign road information includes at least traffic sign information, traffic signal light information, lane line information, road edge information, and road guidance information. The stereo camera, forward long-range millimeter microwave radar and four-corner medium-range millimeter microwave radar are used to receive obstacle information, which includes at least pedestrian information, vehicle information, and other obstacle information. The sensor sends the received traffic sign road information and obstacle information to the automatic driving controller.

The positioning module includes CPS inertial navigation and high-precision maps. GPS inertial navigation receives GPS signals and vehicle sensor (such as vehicle speed and wheel speed) signals, calculates the absolute position information of the own vehicle, and sends the position information to the automatic driving control via Ethernet The high-precision map stores relevant information such as road edges, lane centerlines, tunnels, road section ramps, road section curvature, ramp entrances and exits, and service area entrances and exits, as well as absolute position information of traffic sign positions and landmark buildings, etc. According to the GPS inertial navigation, the high-precision map information in a certain area near the absolute position of the vehicle is calculated through the Ethernet automatic driving controller.

The human-computer interaction module includes instrumentation, in-vehicle display, in-vehicle touch screen, autopilot switch, autopilot status indicator and emergency brake button. Instruments, in-vehicle display screens, and in-vehicle touch screens display vehicle status information, road environment perception information, and information about the automatic driving system during automatic driving; the in-vehicle touch screen can also receive relevant instructions from the driver, such as setting the starting point for automatic driving and destination; the autopilot switch turns on and off the autopilot system, the autopilot status light is on to indicate that it is in autopilot mode, and the emergency brake button is used to stop in an emergency.

The vehicle execution module includes vehicle EMS, EPS, ESC, EPB, BCM and TCU, which are used to control the accelerator opening, steering, braking, parking, gear position, lights and horn according to the control instructions sent by the automatic driving controller.

As shown in Figure 2, the automatic driving controller consists of five parts: the automatic driving perception unit, the vehicle vertical and horizontal motion control module, the exception handling module, the driving mode switching control module and the human-computer interaction control module. The automatic driving perception module includes multi-sensor information fusion, vehicle fusion positioning, global path planning module and local path planning module.

The sensor information fusion module receives, processes and analyzes traffic sign information, traffic signal light information, lane line information, road edge information, road guidance information, pedestrian information, vehicle information, For other obstacle information, after detection and identification, combined with the vehicle location information sent by the vehicle fusion positioning module, multi-source information fusion processing is performed to establish a local environment map for vehicle driving.

The vehicle fusion positioning module receives the positioning information of GPS inertial navigation, and fuses with the vehicle speed, rotation speed, high-precision map and relevant information of the multi-sensor information fusion module to perform vehicle positioning.

The global path planning module receives the start point and end point set by the driver in the human-computer interaction module, and completes the global path planning of the automatic driving behavior including the driving route according to the vehicle position information sent by the vehicle fusion positioning module.

The local path planning module receives the driving route planned by the global path planning module and the positioning information of the vehicle fusion positioning module, and plans a local driving path that meets the traffic rules and safety requirements on the vehicle local environment map established by the multi-sensor information fusion module ;

The vehicle longitudinal and lateral motion control module, according to the vehicle's own posture and local planning path, executes the vehicle module according to the control strategy, including accelerator (EMS), steering (EPS), brake (ESC), parking (EPB), gear ( TCU), light and horn (BCM) issue CAN bus commands for control, and control and adjust through feedback information, and operate the vehicle to realize the vehicle action according to the planned path.

The abnormal processing module monitors the vehicle status, automatic driving controller and road weather conditions, and receives alarm signals, such as (abnormal fuel alarm, tire pressure abnormal alarm, automatic driving controller abnormal alarm, sensor abnormal alarm, weather condition early warning, etc., through the whole The vehicle network will send the alarm message to the controller for processing) If abnormal or unsuitable for automatic driving is found, the alarm will prompt the driver to take over the vehicle, end the automatic driving, and brake to stop if necessary.

The driving mode switching control module monitors the input signal of the human-computer interaction module, and after monitoring the active (automatic driving system switch state, steering wheel movement state, accelerator and brake pedal state) or passive (mainly provided by the exception processing module) driving mode switching signal , control the smooth and safe switching between the automatic driving mode and the manual driving mode, and send the switched vehicle's own attitude information to the vehicle's vertical and lateral motion control module.

The human-computer interaction control module accepts various input commands from the driver of the human-computer interaction module, as well as the obstacle position information and the vehicle position information sent by the automatic driving perception unit, and sends them to the driving mode switching control module through the vehicle network for processing , and send the vehicle status, automatic driving controller status, road environment perception results, and alarm prompts to the human-computer interaction module for display through the vehicle network, helping the driver understand and monitor the system status.

The bus interface of the automatic driving controller adopts Ethernet to connect with GPS inertial navigation, high-precision map and visual sensor, and uses CAN bus to connect with millimeter wave radar, human-computer interaction module, EPS, EMS, ESC, EPB, TCU and BCM.

The automatic driving controller, the software system is developed based on the Ubuntu operating system, and the controller is in the form of an embedded controller, which can be realized by using NVIDIA Tegra K1, X1 and other chips.

The above is only a specific embodiment of the present invention. It should be pointed out that any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention shall be covered by the protection scope of the present invention. within.

Claims (10)

1. a kind of automated driving system based on highway, it is characterised in that including：

Sensor：For obtaining traffic sign road information and obstacle information, and it is sent to auto-pilot controller；

Locating module：The absolute location information and carwash for coming from car for being calculated according to GPS inertial navigations and high-precision cartographic information are exhausted To the high-precision cartographic information near position, and it is sent to auto-pilot controller；

Human-computer interaction module：For receiving driver command, and car status information during display automatic Pilot, road Environment sensing information, automated driving system relevant information；

Auto-pilot controller：For receive sensor transmission traffic sign road information and obstacle information, and with positioning The high-precision cartographic information that module is sent is merged, and vehicle traveling local environment map is set up, according in human-computer interaction module The beginning and end that driver is set, on the vehicle traveling local environment map that multi-sensor information fusion module is set up, rule The local travel route for meeting traffic rules and security requirement is drawn, mould is performed to vehicle finally according to the local travel route Block sends control instruction；

Vehicle performing module：For receiving the control instruction that the auto-pilot controller is sent, corresponding action is performed.

2. the automated driving system as claimed in claim 1 based on highway, it is characterised in that：The sensor includes monocular Camera, the monocular camera is used to receive traffic sign road information, and the traffic sign road information at least includes traffic mark Will information, traffic lights information, lane line information, road edge information, road guidance information.

3. the automated driving system as claimed in claim 1 based on highway, it is characterised in that：The sensor includes solid Camera, forward direction length are away from millimeter microwave radar and corner middle-range millimeter microwave radar, and the stereoscopic camera, the forward direction length are away from millimeter Microwave radar and the corner middle-range millimeter microwave radar are used to receive obstacle information, and the obstacle information at least includes row People's information, information of vehicles, other obstacle informations.

4. the automated driving system as claimed in claim 3 based on highway, it is characterised in that：The monocular camera and solid Camera is installed on room mirror rear.

5. the automated driving system as claimed in claim 3 based on highway, it is characterised in that：The forward direction length is away from millimeter wave Radar is installed on vehicle front bumper center, and the corner middle-range millimetre-wave radar includes preceding left comer middle-range millimetre-wave radar, preceding Right corner middle-range millimetre-wave radar, rear left comer middle-range millimetre-wave radar, rear and right corner middle-range millimetre-wave radar, and be respectively arranged in front and rear The side position of bumper two.

6. the automated driving system as claimed in claim 1 based on highway, it is characterised in that：The auto-pilot controller Unit, vehicle vertically and horizontally action control module, driving model are perceived including being integrated in the automatic Pilot of an editable gate array Switching control module and human-computer interactive control module, the editable gate array include can at least with sensor, locating module The data transmission interface of data interaction is carried out, the control signal coffret of control instruction can be sent to vehicle performing module, with And the Man Machine Interface of signal transmission is carried out with human-computer interaction module；

The automatic Pilot, which perceives unit, to be used to collect traffic sign road information and obstacle information by data transmission interface, Carry out merge generation part run routing information with high-precision map, GPS inertial navigation informations, wheel, speed information, and by the office Portion's run routing information is sent to vehicle vertically and horizontally action control module, is sent by obstacle position information and from truck position information Give human-computer interactive control module；

Vertically and horizontally action control module is used to receive the local driving path vehicle and human-computer interactive control module is sent Vehicle itself attitude information, by control signal coffret to each module of execution unit send control instruction；

The driving model switching control module is used to monitor the input of human-computer interactive control module, control automatic driving mode and people Work driving model switches, and itself attitude information is sent to vehicle vertically and horizontally action control module by the vehicle after switching；

The human-computer interactive control module is used to receive the control instruction that human-computer interaction module is sent by Man Machine Interface, with And automatic Pilot perceives the obstacle position information of unit transmission and is sent to from truck position information, and by Man Machine Interface Human-computer interaction module display output.

7. the automated driving system as claimed in claim 6 based on highway, it is characterised in that：The automatic Pilot perceives single Member includes global path planning module, and the global path planning module is used to receive what driver in human-computer interaction module was set Beginning and end, the vehicle position information sent according to the Vehicle Fusion locating module completes driving automatically comprising travel route Sail behavior global path planning.

8. the automated driving system as claimed in claim 7 based on highway, it is characterised in that：The automatic Pilot perceives single Member includes local paths planning module, and the local paths planning module receives the travel route of global path planning module planning The vehicle position information sent with the Vehicle Fusion locating module, the vehicle set up in multi-sensor information fusion module Travel on local environment map, planning meets the local driving path of traffic rules and security requirement, and by the partial row Sail path and be sent to the vehicle vertically and horizontally action control module.

9. the automated driving system as claimed in claim 6 based on highway, it is characterised in that：The auto-pilot controller Also include exception processing module, the exception processing module is used to monitor vehicle-state, auto-pilot controller and road weather Situation, receives alarm signal and is handled, and the alarm signal is sent into human-computer interactive control module.

10. the automated driving system as claimed in claim 1 based on highway, it is characterised in that：The vehicle performing module Including vehicle EMS, EPS, ESC, EPB, BCM and TCU, the control instruction for being sent according to auto-pilot controller is opened throttle Degree, steering, braking, parking, gear, light and loudspeaker are controlled.