An intelligent vehicle in-out warehouse vehicle speed and trajectory planning method

Abstract

The present invention discloses a method for planning vehicle speed and trajectory during intelligent vehicle entry and exit from a garage, belonging to the field of intelligent vehicle control. Aiming at the scenario of connected vehicles entering and exiting an underground garage, based on the recognition function of a binocular camera and the stored data in the cloud, the power type and related parameters of the current vehicle entering and exiting the garage are obtained. Subsequently, based on the binocular camera and the imu sensor, the vehicle speed and trajectory information of the vehicle entering and exiting the garage are acquired. Thereafter, using neural networks and deep learning, according to the previously obtained vehicle speed and trajectory information, the vehicle speed and trajectory after neural network optimization for the corresponding vehicle type are obtained, so as to facilitate guiding subsequent vehicles to automatically drive in and out of the garage or providing assistance to the driver. The purpose of the present invention is to plan an optimized trajectory and vehicle speed by collecting the vehicle speed and trajectory data of the vehicle entering and exiting the garage, and then assist and guide subsequent vehicles to enter and exit the underground garage to ensure the safety and efficiency of the vehicle entering and exiting the garage.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent vehicle control, and particularly relates to a method for planning the vehicle speed and trajectory of an intelligent vehicle when entering and leaving a garage. Background Art

[0002] With the development of the digital economy and big data technology, intelligent vehicles provide new options for people's travel. When a vehicle enters or exits an underground garage, the driver is often affected by factors such as their own experience, line of sight, and the structure and environment of the underground garage itself. This kind of influence not only reduces the efficiency of the vehicle entering and exiting the underground garage, but also poses a safety hazard at the entrance and exit of the underground garage. Many solutions have been proposed to better achieve the rapid and safe entry and exit of vehicles from underground garages. For example, the invention patent published on March 7, 2023: Publication number: CN115742963A, an auxiliary system for guiding a vehicle to reverse into a garage. This invention proposes a system for assisting a vehicle to reverse into a garage, including a surround-view camera, a display screen, and a controller. The surround-view camera captures the surrounding video, and the controller includes a panoramic image, manual parking space selection, path planning, and driving operation reminder modules. The panoramic image module stitches the surrounding video into a panoramic image, the path planning module plans the reverse parking path based on vehicle positioning and path planning algorithms, and the driving operation reminder module prompts the user. The system improves driving safety and intelligently assists the user to complete reverse parking. However, this invention can only target a single vehicle and does not further propose a plan for vehicle speed. Another example is the invention patent published on May 16, 2023: Publication number: CN116122197A, an early warning device for vehicle entry and exit identification. This invention is more inclined to the identification and obstruction of vehicles and cannot fundamentally automatically plan the trajectory of vehicles entering and exiting the garage. Summary of the Invention

[0003] In order to improve the safety and efficiency of a vehicle when entering and leaving an underground garage, the present invention combines artificial intelligence and big data technologies and proposes a method for planning the vehicle speed and trajectory of an intelligent vehicle when entering and leaving a garage, characterized in that:

[0004] Step 1: Identification of vehicle type and acquisition of vehicle speed and trajectory: First, install a binocular camera at a suitable position at the entrance and exit of the corresponding underground garage. When a vehicle is detected entering or exiting the entrance and exit, the binocular camera identifies the current vehicle and, in combination with the database in the cloud, obtains information such as the brand model, type of power system, power system parameters, and overall vehicle size parameters of the vehicle. Then, the binocular camera identifies the macroscopic trajectory of the whole vehicle and obtains the trajectory of the vehicle's outer contour. At the same time, the IMU sensor installed inside the vehicle obtains the acceleration and angular acceleration of the vehicle during the entire process of entering and leaving the warehouse. By combining the obtained macroscopic trajectory of the vehicle, acceleration, and angular acceleration, the vehicle speed and trajectory parameters of the vehicle during entry and exit can be obtained. Based on the above steps, the relevant parameters of a series of vehicles entering and leaving the garage and the vehicle speed and trajectory parameters are formed into big data information and transmitted to the cloud for storage. The cloud can reasonably classify the current vehicle type according to different classification methods and form a vehicle speed and trajectory database for vehicles of this type entering and leaving the current garage.

[0005] Step 2: Obtain the optimized trajectory of the current vehicle type: Combining the vehicle parameters and vehicle speed and trajectory parameters stored in the cloud, using BP neural network and deep learning, initially form the optimized vehicle speed and optimized trajectory of various types of vehicles in the current garage in the cloud; the trajectories of subsequent vehicles entering and leaving the warehouse will be continuously transmitted to the cloud, and the vehicle speed and trajectory during entry and exit will be gradually improved and optimized; the cloud combines the optimized trajectory and vehicle speed with the type of the current vehicle entering and leaving the warehouse and sends them to the current vehicle. By optimizing the vehicle speed, the total output power of the optimized power source can be determined; by optimizing the trajectory, the output of the steering system can be obtained; thus, the purpose of assisting the vehicle to enter and leave the garage or realizing the entry and exit of an autonomous vehicle into the garage can be achieved.

[0006] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0007] 1. The vehicle entry and exit speed and trajectory planning method described in this patent can effectively utilize big data technology to automatically train the optimized entry and exit vehicle speed and trajectory under specific garage conditions by identifying the vehicle speed and trajectory signals of historical vehicles.

[0008] 2. The vehicle entry and exit speed and trajectory planning method described in this patent can not only assist the driver in operating the vehicle to enter and leave the underground garage, but also reasonably combine communication technology to directly guide the vehicle to automatically enter and leave the underground garage, and can be applied to intelligent vehicles with autonomous driving functions. Description of the Drawings

[0009] The following description of the embodiments will become easier to understand with reference to the drawings, where:

[0010] Figure 1 is the framework of the intelligent vehicle entry and exit speed and trajectory planning method according to the embodiment of the present invention; Detailed implementation manners

[0011] The embodiments of the present invention will be described in detail below. Examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary only for explaining the present invention and should not be construed as limiting the present invention.

[0012] A framework for a method of planning vehicle speed and trajectory for intelligent vehicle entry and exit from a garage will be described below with reference to the accompanying drawings, but the present invention is not limited to these embodiments.

[0013] Refer to the attached Figure 1 The economic vehicle speed planning method based on networked information according to the present invention is divided into two steps. Step 1: Combining the signal light phase, the vehicle's own position, and the driving path provided by the cloud, the reference time for passing through each signal light intersection is obtained through forward solution and backward recursion. Step 2: Using the MPC controller, on the basis of successfully passing through each signal light intersection, the vehicle speed of the vehicle is planned by comprehensively considering the economy and comfort of the vehicle.

[0014] Step 1: Vehicle type recognition and acquisition of vehicle speed and trajectory: First, install a binocular camera at a suitable position at the entrance and exit of the corresponding underground garage. When a vehicle is detected entering or exiting the entrance and exit, the binocular camera recognizes the current vehicle, and in combination with the database in the cloud, information such as the brand model, power system type, power system parameters, and overall vehicle size parameters of the vehicle is obtained. Then, the binocular camera recognizes the macroscopic trajectory of the whole vehicle and obtains the trajectory of the outer contour of the whole vehicle. At the same time, the IMU sensor installed inside the vehicle obtains the acceleration and angular acceleration of the vehicle during the entire entry and exit process. By combining the obtained macroscopic trajectory of the vehicle, acceleration, and angular acceleration, the vehicle speed and trajectory parameters of the vehicle during entry and exit can be obtained. Based on the above steps, the relevant parameters and vehicle speed and trajectory parameters of a series of vehicles entering and exiting the garage are formed into big data information and transmitted to the cloud for storage. The cloud can reasonably classify the current vehicle type according to different classification methods and form a vehicle speed and trajectory database for vehicles of this type entering and exiting the current garage;

[0015] Step 2: Obtain the optimized trajectory of the current vehicle type: Combine the vehicle parameters and vehicle speed trajectory parameters stored in the cloud, and use the BP neural network and deep learning to initially form the optimized vehicle speeds and optimized trajectories of various types of vehicles in the current garage in the cloud; the trajectories of subsequent vehicles entering and leaving the garage will be continuously transmitted to the cloud, and the vehicle speeds and trajectories of entering and leaving the garage will be gradually improved and optimized; the cloud will send the optimized trajectories and vehicle speeds, combined with the type of the current vehicle entering and leaving the garage, to the current vehicle; by optimizing the vehicle speed, the total output power of the optimized power source can be determined; by optimizing the trajectory, the output of the steering system can be obtained; thereby achieving the purpose of assisting the vehicle to enter and leave the garage or enabling an autonomous vehicle to enter and leave the garage.

[0016] In the present invention, terms such as IMU are used more frequently, but the possibility of using other terms is not excluded. These terms are used only to more conveniently describe and explain the essence of the present invention; interpreting them as any additional limitation is contrary to the spirit of the present invention.

[0017] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and purposes of the present invention, and the scope of the present invention is defined by the claims and their equivalents.

Claims

1. An intelligent vehicle in-out warehouse vehicle speed and trajectory planning method, characterized in that: It includes the following steps: Step 1, vehicle type recognition and vehicle speed trajectory acquisition: First, install a binocular camera at a suitable position at the entrance and exit of the corresponding underground garage. When a vehicle is detected entering or exiting the entrance and exit, the binocular camera identifies the current vehicle and, in combination with the database in the cloud, obtains information such as the brand model, power system type, power system parameters, and overall vehicle size parameters of the vehicle. Then, the binocular camera identifies the macroscopic trajectory of the whole vehicle and obtains the trajectory of the vehicle's outer contour. At the same time, the IMU sensor installed inside the vehicle obtains the acceleration and angular acceleration of the vehicle during the entire process of entering and exiting the garage. The cloud combines the obtained macroscopic trajectory, acceleration, and angular acceleration of the vehicle to predict the vehicle speed and trajectory parameters when the vehicle enters and exits the garage. Based on the above steps, the relevant parameters and vehicle speed trajectory parameters of a series of vehicles entering and exiting the garage are formed into big data information and transmitted to the cloud for storage. The cloud can reasonably classify the current vehicle type according to different classification methods and form a vehicle speed and trajectory database for vehicles of this type entering and exiting the current garage; Step 2, obtaining the optimized trajectory of the current vehicle type: Combining the vehicle parameters and vehicle speed trajectory parameters stored in the cloud, using BP neural network and deep learning, initially form the optimized vehicle speed and optimized trajectory of each type of vehicle in the current garage in the cloud. The trajectories of subsequent vehicles entering and exiting the garage will be continuously transmitted to the cloud, and the vehicle speed and trajectory of entering and exiting the garage will be gradually improved and optimized; The cloud sends the optimized trajectory and vehicle speed, combined with the type of the current vehicle entering and exiting the garage, to the current vehicle; By optimizing the vehicle speed, the total output power of the optimized power source can be determined; by optimizing the trajectory, the output of the steering system can be obtained; thus, the purpose of assisting the vehicle to enter and exit the garage or realizing the entry and exit of an autonomous vehicle from the garage can be achieved.

Images