CN107200020A - It is a kind of based on mix theory pilotless automobile self-steering control system and method - Google Patents

Abstract

The invention discloses an autonomous steering control system and method for an unmanned vehicle based on hybrid theory, which belongs to the field of lateral control of unmanned vehicles. Planning module, decision-making control module, and manipulation execution module. The present invention also provides an autonomous steering control method for unmanned vehicles based on the hybrid theory. By introducing the hybrid control theory, the control problem of the steering system under complex working conditions is transformed into a multi-mode control and a switching coordination problem between control algorithms. Solve the problem, realize the control of the autonomous steering control system of unmanned vehicles under all working conditions, and introduce appropriate control algorithms in each working condition, which can not only meet the local control performance of the system, but also achieve the purpose of overall optimization, and realize the best unmanned steering control system. Steering control performance of human-driven cars, improving the handling stability and intelligence level of unmanned vehicles.

Description

A self-driving car autonomous steering control system and method based on hybrid theory

technical field

The invention relates to the field of lateral control of unmanned vehicles, in particular to an autonomous steering control system and method for unmanned vehicles based on hybrid theory.

Background technique

The major transformation of automobile technology centered on the three major technologies of power electrification, lightweight structure and intelligent driving is being carried out in depth on a global scale. In 2015, the State Council officially released "Made in China 2025", which listed "energy-saving and new energy vehicles" as one of the ten key breakthrough development areas, and generally pointed out the development direction and This is the first time that intelligent networked vehicles have been promoted to the level of national strategy. The Ministry of Industry and Information Technology further clarified: by 2020, it is necessary to master the overall technology of intelligent assisted driving and various key technologies, and initially establish an independent research and development system and production supporting system for intelligent networked vehicles; by 2025, it is necessary to master the overall technology of autonomous driving and various key technologies , Establish a relatively complete independent research and development system, production supporting system and industrial cluster for intelligent networked vehicles.

Unmanned vehicles refer to equipped with advanced on-board sensors, controllers, actuators and other devices, and integrate modern communication and network technologies to realize the exchange and sharing of intelligent information between vehicles and X (people, vehicles, roads, background, etc.), and have complex environments. Functions such as perception, intelligent decision-making, collaborative control, and execution can realize safe, comfortable, energy-saving, and efficient driving, and can eventually replace a new generation of human-operated vehicles.

Facing the ultimate goal of unmanned driving, the intelligent vehicle lateral control system is required to have accurate, efficient and reliable control capabilities under multiple working conditions to ensure vehicle steering stability, driving safety and ride comfort. The traditional single steering control algorithm It cannot solve and coordinate the control requirements and control functions of the autonomous steering control system under different working conditions.

From the perspective of the safety and stability of unmanned vehicle control, different working conditions should have different control objectives and emphases, so that the overall comprehensive performance can be optimized. Autonomous steering control of unmanned vehicles essentially has discrete events. The coexistence of continuous dynamics and the interaction between discrete events and continuous dynamic behaviors show the hybrid system characteristics of typical switching control. Hybrid control system theory has better performance than traditional methods, and can obtain better results than continuous dynamic or discrete event dynamic systems alone, and solve complex problems that traditional controllers cannot solve.

Contents of the invention

In order to solve the above-mentioned technical problems, the present invention provides an autonomous steering control system and method for unmanned vehicles based on the hybrid theory, and introduces the hybrid control system theory to analyze the hybrid characteristics of the steering control system to realize the hybridization of the steering control system of unmanned vehicles. Control can improve the control capability of the steering control system of unmanned vehicles and meet the driving stability requirements of autonomous driving in multiple working conditions.

An autonomous steering control system for unmanned vehicles based on hybrid theory, including an environment perception module, a vehicle's own state information module, a path planning module, a decision-making control module, and a manipulation execution module.

The environment perception module uses vision, radar and positioning systems to perceive the external environment information of the vehicle;

The vehicle's own state information module uses inertial devices and CAN bus to obtain the running state information of the vehicle and estimates the driving parameters; the vehicle's own state information module uses the automobile human-computer interaction unit to prompt the driver whether there is a fault in the control system.

The path planning module performs global and local path planning based on the information transmitted by the environment perception module, based on the absolute position of the vehicle in the map system and the relative position of the vehicle, surrounding obstacles, and lane lines, according to the path with the least energy consumption or the shortest path length. Evaluation criteria, looking for a collision-free vehicle expected travel path.

The decision-making control module includes: an upper control unit and a lower control unit, the hybrid controller belongs to the upper control unit, and the execution motor belongs to the lower control unit; the hybrid controller includes: a steering controller, a switching supervisor, a switching controller, a stability supervision The upper control is used to output the front wheel rotation angle and yaw angular velocity value at each moment, and the output value of the upper control is passed to the lower control as the control quantity, and the lower control realizes the rotation angle and yaw of the vehicle at each moment by controlling the execution motor angular velocity;

The steering controller is designed to meet the control objectives of the low-speed, medium-speed and high-speed working conditions, so as to meet the control requirements of the vehicle under different working conditions;

The switching supervisor drives the hybrid controller module to perform effective mode switching according to the change of vehicle speed and whether there is a fault in the autonomous steering control system, while ensuring the stability of the system during the switching process;

According to the working mode of the unmanned vehicle recognized by the switching supervisor, the switching controller selects the appropriate control algorithm in the steering controller to realize the switching coordination between different control modes and control strategies of the unmanned vehicle.

The stability supervisor is used to monitor the unstable characteristic quantities under each control mode and its corresponding control algorithm in real time, identify its unstable trend, and forcibly limit its output amplitude to ensure the global bounded stability of the system.

The manipulation execution module is used for performing autonomous steering of the unmanned vehicle.

In the above solution, the hybrid controller module includes four modes, namely, low-speed mode m1, medium-speed mode m2, high-speed mode m3, and system failure mode m4;

In the above solution, the low-speed mode m1, the medium-speed mode m2, and the high-speed mode m3 belong to the controlled state, and the system failure mode m4 belongs to the uncontrolled state.

A self-driving car autonomous steering control method based on hybrid theory, comprising the following steps:

Step 1), the environment perception module perceives the external environment information of the vehicle, and sends the obtained information to the path planning module;

Step 2), the path planning module performs global and local path planning based on the information obtained by the environment perception module, based on the absolute position of the vehicle in the map system and the relative position of the vehicle to surrounding obstacles and lane lines;

Step 3), the decision-making control module converts the result of path planning into an execution command for trajectory tracking according to the dynamic state and running state of the vehicle obtained by the vehicle's own information module, and realizes the control of the lateral position and heading angle of the vehicle through the manipulation execution module. Track the desired travel path. The position deviation and heading deviation are used as the input of the steering controller in the hybrid controller in the upper control, and the switching supervisor determines the current working mode according to the vehicle speed and whether there is a fault in the autonomous steering control system, and ensures the stability of switching between modes , the switching controller selects the most suitable control algorithm in the steering controller under the different modes recognized by the switching supervisor. The front wheel angle and yaw rate are the final outputs of the hybrid controller. As the control quantity, the front wheel rotation angle and yaw rate of the vehicle are realized by controlling the actuator motor.

Step 4), the manipulation execution module drives the manipulation actuator according to the rotation direction and output torque of the execution motor output by the decision-making control module, and executes the autonomous steering of the unmanned vehicle, so that the unmanned vehicle can track the desired trajectory;

The above method also includes: when a system failure occurs in the autonomous steering control system, the driver will be prompted through the human-computer interaction unit in the vehicle's own information module, and the driver will immediately switch the vehicle driving mode to the human driving mode.

The beneficial effects of the present invention are:

1. The present invention introduces the hybrid system theory, transforms the operation and control problems of the steering system under complex working conditions into multi-mode control and the switching coordination problem between control algorithms to solve, and uses the vehicle speed as the basis for the division of different modes of the autonomous steering control system , decompose the complex lateral control into several synthetic controls under single working conditions, build a "hybrid switching model system" for the steering control system of unmanned vehicles, and realize the modeling and control of the hybrid dynamic system of autonomous steering of unmanned vehicles .

2. The present invention introduces appropriate control algorithms in different working conditions of the autonomous steering control system to meet the control requirements and control functions under different working conditions, and also meet the control requirements in the overall operation process, improving the steering performance and handling of unmanned vehicles stability.

3. Establish a switching supervisor and a stability supervisor of the autonomous steering control system to ensure the gradual stability and global bounded stability of the system during the switching process, so as to not only meet the local stability of the system, but also achieve the purpose of overall optimization and system stability, and improve the Stability of autonomous steering control for driverless cars.

Description of drawings

Figure 1 is a schematic structural diagram of an autonomous steering control system for an unmanned vehicle;

Figure 2 is a block diagram of the hybrid controller for autonomous steering multi-mode switching control of unmanned vehicles.

detailed description

The following will be described in detail in conjunction with the accompanying drawings of the present invention.

As shown in Figure 1, an autonomous steering control system for unmanned vehicles based on hybrid theory, including an environment perception module, a vehicle state information module, a path planning module, a decision-making control module, and a manipulation execution module;

The environmental perception module includes a camera, radar system, inertial navigation/GPS positioning system, and information processing system. The camera is installed on the front and rear windshields of the vehicle to collect information such as lane lines, signs, traffic lights, etc.; the radar system is used to detect obstacles; The inertial navigation/GPS positioning system is used to obtain vehicle attitude parameters, speed and position information; the information processing system is used to process the information obtained by the camera, radar system, and inertial navigation/GPS positioning system.

The vehicle's own state information module includes inertial devices, CAN bus, and human-computer interaction unit; the inertial device specifically refers to inertial navigation, which uses accelerometer and gyroscope sensors to measure vehicle parameters; inertial devices and CAN bus are used to obtain the running status of the vehicle information and estimate the driving parameters; due to the direct measurement of some key states in the driving process of the car, there are problems such as high measurement cost and difficulty in accurate measurement. Therefore, on the basis of the vehicle attitude information received by the inertial navigation system, the estimation method is used to obtain the complete vehicle state parameters such as lateral velocity, yaw rate, and center of mass side slip angle with the help of the vehicle dynamics model, so as to accurately obtain the state parameters of the vehicle during driving. The state parameters can match the steering control strategy with the controlled object and achieve good control performance; the human-computer interaction unit will prompt the driver when the autonomous steering control system fails.

The path planning module performs global path planning and local path planning based on the relative position of the vehicle and surrounding obstacles, lane lines, and the relative position in the map system, and finds a collision-free vehicle expectation according to the evaluation criteria of the least energy consumption or the shortest path length driving path.

The decision-making control module includes an upper control unit and a lower control unit; the hybrid controller belongs to the upper control unit, and the execution motor belongs to the lower control unit; the hybrid controller includes: a steering controller, a switching supervisor, and a switching controller; the upper control is used to output each The front wheel angle and yaw rate value at each moment, the output value of the upper control is sent to the lower control as the control quantity, and the lower control realizes the front wheel angle and yaw rate of the vehicle at each moment by controlling the executive motor; the decision-making control module according to Judgment of steering control mode and formulation of control strategy based on the results of environment perception and path planning, the status information of the vehicle and each subsystem, and the feedback status information of the actuator;

Steering controllers design controllers that meet the control objectives for low-speed, medium-speed, and high-speed operating conditions, and introduce appropriate control algorithms in low-speed, medium-speed, and high-speed modes to meet the control requirements of the vehicle under different operating conditions;

The switching supervisor switches the working mode of the hybrid controller according to the vehicle speed obtained by the speed sensor in the vehicle's own state information module and whether the autonomous steering control system provided by the human-computer interaction module is faulty, and at the same time supervises the control modes of the unmanned vehicle The transition process between them reduces the impact and oscillation caused by the mode switching during the steering control execution process, so as to achieve the gradual stability of the switching process; it is proposed to use 20km/h as the medium and low speed switching point, and 60km/h as the medium and high speed switching point. The vehicle speed is between 0-20km/h for low-speed mode, between 20-60km/h for medium-speed mode, and above 60km/h for high-speed mode;

The switching controller selects the appropriate control algorithm in the steering controller according to the working mode of the unmanned vehicle identified by the switching supervisor, so as to realize the switching coordination between different control modes and control strategies of the unmanned vehicle.

The stability supervisor is used to monitor the unstable characteristic quantities under each control mode and its corresponding control algorithm in real time, identify its unstable trend, and forcibly limit its output amplitude to ensure the global bounded stability of the system.

The manipulation execution module is used to perform the autonomous steering of the unmanned vehicle.

The hybrid controller module contains four modes, which are low-speed mode m1, medium-speed mode m2, high-speed mode m3, and system failure mode m4; low-speed mode m1, medium-speed mode m2, and high-speed mode m3 belong to the controlled state, and system failure mode m4 belongs to the state of no control.

A self-driving car autonomous steering control method based on hybrid theory, comprising the following steps:

Step 1), the environment perception module perceives the external environment information of the vehicle, and sends the obtained information to the path planning module;

Step 2), the path planning module performs global and local path planning based on the information obtained by the environment perception module, based on the absolute position of the vehicle in the map system and the relative position of the vehicle to surrounding obstacles and lane lines, according to the least energy consumption or the shortest path length The evaluation standard, looking for a collision-free vehicle expected driving path;

Step 3), the decision-making control module converts the result of path planning into an execution command for trajectory tracking according to the dynamic state and running state of the vehicle obtained by the vehicle's own information module, and realizes the control of the lateral position and heading angle of the vehicle through the manipulation execution module. Track the desired travel path. The position deviation and heading deviation are used as the input of the steering controller in the hybrid controller in the upper control, and the switching supervisor determines the current working mode according to the vehicle speed and whether there is a fault in the autonomous steering control system, and ensures the stability of switching between modes , the switching controller selects the most suitable control algorithm in the steering controller under the different modes identified by the switching supervisor, and the stability supervisor is used to monitor the unstable characteristic quantities of each control mode and its corresponding control algorithm in real time, and identify its unstable trend , to forcefully limit its output amplitude and ensure the global bounded stability of the system. The front wheel angle and yaw rate are the final output of the hybrid controller, and the lower control takes the front wheel angle and yaw rate as the control quantity, and realizes the front wheel angle and yaw rate of the vehicle by controlling the executive motor;

The control algorithm used in each working mode is as follows:

1) In the low-speed mode m1, the driving speed of the vehicle is low and the safety is high, while the PID control algorithm has the advantages of simple control algorithm, separate application, easy parameter adjustment, and good control effect. Therefore, the low-speed mode can use, for example, PID Adaptive control algorithm.

2) In the medium-speed mode m2, under medium-speed conditions, vehicles mostly drive on urban roads, the traffic is complex, and the steering control accuracy is high. Therefore, in the medium-speed mode, for example, an optimal control algorithm can be used.

3) In the high-speed mode m3, the driving speed of the vehicle is relatively high, and the model predictive control algorithm has unique advantages in solving the trajectory tracking control problem of unmanned vehicles at high speeds. Therefore, in the high-speed mode, for example, model-based prediction can be used control algorithm.

4) In the system failure mode m4, the driver switches the driving mode to manual control, and does not need to perform autonomous steering control.

Step 4), the manipulation execution module drives the manipulation actuator according to the rotation direction and output torque of the execution motor output by the decision-making control module, and executes the autonomous steering of the unmanned vehicle, so that the unmanned vehicle can track the desired trajectory;

The above method also includes: when a system failure occurs in the autonomous steering control system, the driver will be prompted through the human-computer interaction unit in the vehicle's own information module, and the driver will immediately switch the vehicle driving mode to the human driving mode.

The above is a detailed introduction to the self-driving car autonomous steering control system and method based on the hybrid theory provided by the present invention. The above is only a preferred embodiment of the present invention, and is only used to illustrate the design ideas and characteristics of the present invention. , is not intended to limit the present invention, and any modification, equivalent replacement, improvement, etc. made under the technical idea of the present invention shall be included within the protection scope of the present invention.

Claims (9)

1. An autonomous steering control system for unmanned vehicles based on hybrid theory, characterized in that it includes: an environment perception module, a vehicle state information module, a path planning module, a decision-making control module, and a manipulation execution module; The environment perception module uses vision, radar and positioning system to perceive the external environment information of the vehicle; The vehicle's own state information module uses inertial devices and CAN bus to obtain the running state information of the vehicle and estimate the driving parameters; The path planning module performs global and local path planning based on the information transmitted by the environment perception module, based on the absolute position of the vehicle in the map system and the relative position of the vehicle, surrounding obstacles, and lane lines, according to the path with the least energy consumption or the shortest path length. Evaluation criteria, looking for a collision-free vehicle expected driving path; The decision-making control module includes: an upper control unit and a lower control unit; the upper control unit is used to output the front wheel rotation angle and the yaw rate value at each moment, and the output value of the upper control unit is passed to the lower control unit as a control quantity, The lower control unit realizes the turning angle and yaw rate of the vehicle at each moment; The manipulation execution module is used for performing autonomous steering of the unmanned vehicle. 2. A kind of self-driving car autonomous steering control system based on hybrid theory according to claim 1, is characterized in that, described host control unit comprises hybrid controller; Described hybrid controller comprises: steering controller, switch supervisor, switching controller, stabilization supervisor; The steering controller is designed to meet the control objectives of the controllers for low-speed, medium-speed and high-speed working conditions, so as to meet the control requirements of the vehicle under different working conditions; The switching supervisor drives the hybrid controller module to perform effective mode switching according to the change of vehicle speed and whether there is a fault in the autonomous steering control system, while ensuring the stability of the system during the switching process; The switching controller selects an appropriate control algorithm in the steering controller according to the working mode of the unmanned vehicle identified by the switching supervisor, so as to realize the switching coordination between different control modes and control strategies of the unmanned vehicle; The stability supervisor is used for real-time monitoring of unstable characteristic quantities under each control mode and its corresponding control algorithm, identifying its unstable trend, and forcibly limiting its output amplitude, so as to ensure the global bounded stability of the system. 3. A kind of self-driving car autonomous steering control system based on hybrid theory according to claim 1, it is characterized in that, described lower control unit comprises executive motor; Described lower control unit realizes vehicle each by controlling executive motor Turning angle and yaw rate at time. 4. a kind of self-driving car autonomous steering control system based on hybrid theory according to claim 2, is characterized in that, described hybrid controller module comprises four kinds of modes, is respectively low-speed mode m1, medium-speed mode m2, High-speed mode m3, system failure mode m4. 5. a kind of self-driving car autonomous steering control system based on hybrid theory according to claim 4, is characterized in that, described low-speed mode m1, medium-speed mode m2, high-speed mode m3 belong to control state, and described system Failure mode m4 belongs to no control state. 6. A kind of self-driving car autonomous steering control system based on hybrid theory according to claim 1, it is characterized in that, described vehicle self state information module can also utilize automobile man-machine interaction unit to prompt driver whether control system exists Fault. 7. A self-driving car steering control method based on hybrid theory, characterized in that, comprising the following steps: Step 1), the environment perception module perceives the external environment information of the vehicle, and sends the obtained information to the path planning module; Step 2), the path planning module performs global and local path planning based on the information obtained by the environment perception module, based on the absolute position of the vehicle in the map system and the relative position of the vehicle to surrounding obstacles and lane lines; Step 3), the decision-making control module converts the result of path planning into an execution command for trajectory tracking according to the dynamic state and running state of the vehicle obtained by the vehicle's own information module, and realizes the control of the lateral position and heading angle of the vehicle through the manipulation execution module. Track the expected driving path; where the position deviation and heading deviation are used as the input of the steering controller in the hybrid controller in the upper control, and the switching supervisor determines the current working mode according to the vehicle speed and whether there is a fault in the autonomous steering control system, and ensures the relationship between each mode. The switching controller selects the most suitable control algorithm in the steering controller under the different modes recognized by the switching supervisor. The front wheel angle and yaw rate are used as the final output of the hybrid controller. The lower control controls the front wheel The steering angle and yaw rate are used as the control quantity, and the front wheel rotation angle and yaw rate of the vehicle are realized by controlling the actuator motor; Step 4), the maneuvering execution module drives the steering actuator according to the rotation direction and output torque of the executive motor output by the decision-making control module, and executes the autonomous steering of the unmanned vehicle, so that the unmanned vehicle can track the desired trajectory. 8. a kind of self-driving car autonomous steering control method based on hybrid theory according to claim 7, is characterized in that, the control algorithm that adopts under each working mode in described step 3) is as follows: 1) In the low-speed mode m1, the driving speed of the vehicle is low and the safety is high. In this low-speed mode, the PID adaptive control algorithm is adopted; 2) In the medium-speed mode m2, the vehicles mostly drive on urban roads, the traffic is complex, and the steering control precision is high, so the optimal control algorithm is adopted in the medium-speed mode; 3) In the high-speed mode m3, the driving speed of the vehicle is relatively high, and the model-based predictive control algorithm is adopted in this high-speed mode; 4) In the system failure mode m4, the driver switches the driving mode to manual control, and does not need to perform autonomous steering control. 9. A kind of self-driving car autonomous steering control method based on confounding theory according to claim 7, it is characterized in that, described method also comprises: when autonomous steering control system breaks down, will pass vehicle own information module The human-computer interaction unit in the middle prompts the driver, and the driver immediately switches the driving mode of the vehicle to the human driving mode.