KR-20260062207-A - METHOD AND DEVICE FOR CONTROLLING AUTONOMOUS DRIVING BASED ON CONTEXT INFORMATION

KR20260062207AKR 20260062207 AKR20260062207 AKR 20260062207AKR-20260062207-A

Abstract

A method and apparatus for autonomous driving control based on contextual information are disclosed. The method comprises: a step of confirming destination information for a destination to be moved to by a user's vehicle; a step of confirming autonomous driving performance information including information indicating the autonomous driving performance of the vehicle; a step of setting a global path to the destination based on the destination information; a step of generating a first driving scenario for the global path and confirming environmental information based on the first driving scenario and traffic condition information regarding traffic volume on the global path; a step of setting a local path by reflecting the autonomous driving performance information, information on the global path, the environmental information, and the traffic condition information; a step of configuring a second driving scenario by reflecting the local path in the first driving scenario and confirming an autonomous driving execution module including a plurality of sub-models corresponding to the second driving scenario; and a step of controlling the autonomous driving of the vehicle using the autonomous driving execution module including the plurality of sub-models.

Inventors

홍승환

Assignees

주식회사 카카오모빌리티

Dates

Publication Date: 20260507
Application Date: 20241028

Claims (14)

In an autonomous driving control method based on context information, A step of verifying destination information for the destination to be reached via the user's mobile device; A step of verifying autonomous driving performance information including information indicating the autonomous driving performance of the above-mentioned mobile body; A step of setting a global path to the destination based on the above destination information; A step of generating a first driving scenario for the above global route, and verifying environmental information based on the first driving scenario and traffic condition information regarding traffic volume on the above global route; A step of setting a local route by reflecting the above autonomous driving performance information, the above global route information, the above environment information, and the above traffic condition information; A step of configuring a second driving scenario by reflecting the local path in the first driving scenario, and identifying an autonomous driving execution module including a plurality of sub-models corresponding to the second driving scenario; and A step comprising controlling the autonomous driving of the mobile body using an autonomous driving execution module including the plurality of sub-models above. Autonomous driving control method.
In Article 1, The step of setting the above global path is, A step of identifying a user movement pattern or preference corresponding to the user's destination information using a global path learning model that has learned the user movement pattern or preference based on the user's driving data and location information; and A step comprising setting the global path that reflects the user's movement pattern or preference corresponding to the user's destination information. Autonomous driving control method.
In Article 2, The step of verifying the above environmental information and traffic situation information is, A step of collecting information collected through a plurality of sensor units provided in the above-mentioned mobile body; and The method includes the step of converting the collected information into text and constructing context information using the text-based information. Autonomous driving control method.
In Article 1, The step of constituting the above context information is, A step comprising inputting the above textualized information into a chain-of-thought analysis model to output a path plan based on a driving scenario. Autonomous driving control method.
In Article 1, The step of constituting the above context information is, A step comprising inputting the above-mentioned textual information into a language model and verifying information regarding traffic reports, accident updates, and changes in road conditions that is analyzed in the form of natural language and output through the language model. Autonomous driving control method.
In Article 1, The above environmental information External source data including at least one of data regarding road congestion information, accident occurrence information, etc. provided by an Intelligent Transport System (ITS), and weather data provided by a weather information system, etc. High-definition map (HDMap) data including at least one of lane locations, traffic lights, road signs, and intersection configurations, and At least one of driving data comprising a node of the above-mentioned precise map data or a link connecting the node, and a polygon indicating a road or lane area. Autonomous driving control method.
In Article 6, The step of verifying the above environmental information and traffic situation information is, The step of configuring the above external source data, precision map data, and driving data into multidimensional data; The above multidimensional data is converted into data in a multi-layer vector format; and A step of connecting the data in the above-described converted multi-layer vector format by defining mutual topological relationships; including Autonomous driving control method.
In Article 1, The step of setting the above regional path is, A step of integrating the above autonomous driving performance information, the above global path information, the above environment information, and the above traffic situation information; Based on the above integrated information, a step of generating at least one candidate region path; and A step of selecting one of the above-mentioned at least one candidate local path and finally determining the selected path as the local path; including, Autonomous driving control method.
In Article 1, The step of finally determining the above-mentioned selected path as a local path is, Among the above environmental information and the above traffic condition information, the method includes the step of identifying dynamic event information and selecting one of the at least one candidate local route based on the dynamic event information. Autonomous driving control method.
In Article 1, A step of requesting at least one of the plurality of sub-modules corresponding to the second driving scenario to a local server or central server connected to the above-mentioned mobile body; A step of receiving at least one of the requested plurality of submodules from a local server or a central server connected to the mobile body; and The method further comprises the step of updating at least one of the plurality of sub-modules to the autonomous driving execution module and loading it onto the vehicle. Autonomous driving control method.
In Article 10, The step of requesting at least one of the plurality of submodules above is, A step of inputting the above environmental information and the above traffic condition information into a metadata recording model, and verifying the metadata output from the above metadata recording model; and A step of requesting at least one of the plurality of submodules using the above-identified metadata Autonomous driving control method.
In Article 1, The method further comprises the step of inputting the autonomous driving performance information, the environment information, and the traffic situation information into a generative learning model to generate a virtualized driving situation, and collecting virtual environment information and virtual traffic situation information corresponding to the generated virtualized driving situation. Autonomous driving control method.
In Article 12, The step of setting the above regional path is, A step of setting a local route by further reflecting the above-mentioned virtual environment information and virtual traffic condition information Autonomous driving control method.
In an autonomous driving control device based on context information, A communication unit that exchanges data with a mobile body; Memory for storing at least one instruction; and It includes a processor that executes at least one instruction stored in the memory using the above data, The above processor is, Check destination information for the destination you wish to travel to via the user's vehicle, and Check autonomous driving performance information including information indicating the autonomous driving performance of the above-mentioned mobile body, and Based on the above destination information, a global path to the above destination is set, and A first driving scenario for the above global route is generated, and environmental information based on the first driving scenario and traffic condition information regarding traffic volume on the above global route are verified, A local route is established by reflecting the above autonomous driving performance information, the above global route information, the above environment information, and the above traffic condition information, and A second driving scenario is configured by reflecting the local path in the first driving scenario, and an autonomous driving execution module including a plurality of sub-models corresponding to the second driving scenario is identified. An autonomous driving control device configured to control the autonomous driving of the vehicle using an autonomous driving execution module including the plurality of sub-models above.

Description

Method and Device for Controlling Autonomous Driving Based on Context Information The present disclosure relates to a method and apparatus for autonomous driving of a mobile body, and more specifically, to a method and apparatus for providing an autonomous driving module and data suitable for a driving environment. Autonomous vehicles are being developed in various mobility sectors, such as vehicles, robots, unmanned mobile devices, and drones, and their commercialization is being sought. As the applications of autonomous driving expand, the demand for accurate and flexible map information is increasing. Providing or generating map information suitable for various devices, ranging from unmanned drones to autonomous vehicles, individually entails significant difficulties, and this diversity can lead to inefficiencies and operational fragmentation. In addition, the autonomous vehicle establishes a driving plan and performs a decision-making process by referring to a pre-set path on a map, and the autonomous vehicle can predict the expected paths of surrounding objects by referring to their movement paths. However, if the actual movement paths of the vehicle and surrounding objects differ from the pre-set paths, the vehicle may malfunction or cause an accident. To prevent malfunctions or accidents involving moving vehicles, autonomous driving systems are developing perception, judgment, and control software, as well as precision maps and training data, through various tests, and installing these components within autonomous driving computing devices to perform services. For the operation of autonomous vehicles in irregular and complex urban environments involving traffic accidents and road construction, providing software suitable for each driving situation is essential; however, current systems operate in isolation and have limitations in that they can only drive under specific conditions. This problem arises because, in unstructured driving environments, autonomous driving modules must be trained using limited edge case data; since it is difficult to secure sufficient such edge case data, it is not easy for the autonomous driving module to learn how to respond to edge cases. FIG. 1 is a diagram illustrating an example of a mobile body communicating with a server and other devices to transmit and receive data. FIG. 2 is a block diagram of a moving body exemplified in the present disclosure. FIG. 3 is a block diagram of a local server according to one embodiment of the present disclosure. FIG. 4 is a block diagram of a central server according to one embodiment of the present disclosure. FIG. 5 is a block diagram showing the configuration of an autonomous driving module according to one embodiment of the present disclosure. FIG. 6 is a diagram illustrating a driving scenario set in a chain accident analysis model equipped in an autonomous driving module according to one embodiment of the present disclosure. FIGS. 7A and 7B are drawings illustrating the operation of providing an autonomous driving module of a central server and a local server equipped in an autonomous driving system according to one embodiment of the present disclosure. Hereinafter, embodiments of the present disclosure are described in detail with reference to the attached drawings so that those skilled in the art can easily implement them. However, the present disclosure may be embodied in various different forms and is not limited to the embodiments described herein. In describing the embodiments of the present disclosure, if it is determined that a detailed description of known configurations or functions could obscure the essence of the present disclosure, such detailed description is omitted. Additionally, parts of the drawings unrelated to the description of the present disclosure have been omitted, and similar parts are denoted by similar reference numerals. In the present disclosure, when a component is described as being "connected," "combined," or "joined" with another component, this may include not only a direct connection but also an indirect connection in which another component exists in between. Furthermore, when a component is described as "comprising" or "having" another component, this means that, unless specifically stated otherwise, it does not exclude the other component but may include additional components. In the present disclosure, terms such as first, second, etc. are used solely for the purpose of distinguishing one component from another component and do not limit the order or importance of the components unless specifically stated otherwise. Accordingly, within the scope of the present disclosure, a first component in one embodiment may be referred to as a second component in another embodiment, and likewise, a second component in one embodiment may be referred to as a first component in another embodiment. In this disclosure, distinct components are intended to clearly describe their respective features and do no