CN-121999626-A - Can exercise regional ODD platform device based on high in clouds-car end is cooperated

CN121999626ACN 121999626 ACN121999626 ACN 121999626ACN-121999626-A

Abstract

The invention provides a exercisable area ODD platform device based on cloud-vehicle end cooperation. The cloud-vehicle-end-collaboration-based exercisable area ODD platform device comprises a cloud core layer, a vehicle-end execution layer and a tool supporting layer, wherein the cloud core layer is used for storing, processing, version management and issuing non-ODD area data, the vehicle-end execution layer is integrated in an intelligent driving vehicle-mounted system and used for data receiving, real-time computing, mode switching and early warning prompt, and the tool supporting layer is used for providing operation, maintenance and verification tools for the cloud core layer and the vehicle-end execution layer. The exercisable area ODD platform device based on cloud-vehicle end coordination provided by the invention has the advantages of realizing the full-flow closed-loop management from acquisition, processing, updating to execution and feedback of non-ODD area data by constructing a three-layer coordination architecture.

Inventors

CAI KEFANG
CHEN HUBIN
TANG RONGJIANG
CHEN ZHENGXIONG

Assignees

桂林电子科技大学

Dates

Publication Date: 20260508
Application Date: 20260206

Claims (7)

1. The device is characterized by comprising a cloud core layer, a vehicle end execution layer and a tool supporting layer which are operated cooperatively; the cloud core layer is used for storing, processing, version management and issuing of non-ODD region data, and comprises: The cloud database is used for storing structured data of a non-ODD region, wherein the structured data at least comprises a high-precision map version number, an associated lane ID, an ODD type, an abnormal starting point and an abnormal ending point; The data processing and updating module is used for executing the new adding, deleting and deduplication logic of the non-ODD area data; The version management and issuing module is used for issuing the non-ODD area data to the vehicle-end execution layer in a short-term OTA updating or long-term cloud platform updating mode; The vehicle end execution layer is integrated in the intelligent driving vehicle-mounted system and is used for data receiving, real-time calculation, mode switching and early warning prompt, and the vehicle end execution layer comprises: The data storage module is used for storing the non-ODD region data issued from the cloud core layer; the real-time sensing and calculating module is used for inquiring the data of the non-ODD area in real time and calculating the distance between the vehicle and the target non-ODD area according to the current position and the navigation path of the vehicle; the decision and execution module is used for executing multi-stage early warning prompt and driving mode switching according to the distance and the ODD type; The data returning module is used for returning the vehicle operation data to the cloud core layer when the preset conditions are met; The tool supporting layer provides operation and maintenance and verification tools for the cloud core layer and the vehicle end execution layer and at least comprises a problem screening tool, an ODD data manufacturing tool, a simulation verification tool and a visual operation and maintenance tool.
2. The cloud-vehicle-end collaboration-based exercisable area ODD platform device according to claim 1, wherein the deduplication logic of the data processing and updating module is configured to automatically discard the newly added data when the difference between the coordinates of the abnormal start point and the abnormal end point of the newly added data and the coordinates recorded in the cloud database is less than or equal to 10 meters.
3. The cloud-vehicle-side collaboration-based exercisable area ODD platform device of claim 1, wherein the execution logic of the decision and execution module comprises: Triggering a human-computer interaction interface prompt when the vehicle is 500 meters away from the starting point of the non-ODD area; triggering a voice broadcasting prompt when the vehicle is 200 meters away from the starting point of the non-ODD area; and after the vehicle enters the non-ODD area, switching the high-order auxiliary driving function into a pure perception mode or degrading the high-order auxiliary driving function into a lane centering control mode according to the ODD type.
4. The cloud-vehicle end collaboration-based exercisable area ODD platform device of claim 3, wherein the execution logic of the decision and execution module further comprises forcing a perceived degradation when detecting that the front 1 km ramp is a ramp, and suppressing an activation instruction of a high-order auxiliary driving function in the perceived degradation area.
5. The cloud-vehicle-end-collaboration-based exercisable area ODD platform device according to claim 1, wherein the preset condition of the data backhaul module is that a high-order auxiliary driving function is switched from an activated state to a deactivated state, and the number of abnormal signals that a driver steps on a brake and detects is equal to or greater than 2.
6. The cloud-vehicle-end collaboration-based exercisable area ODD platform device according to claim 1, wherein the ODD data creation tool in the tool support layer is configured to input a navigation route, an ODD type and a start-end point coordinate, automatically generate an associated lane ID and a shape point by calling a map open platform interface, and calculate offsets of an abnormal start point and an abnormal end point to generate a standardized non-ODD area data file.
7. The cloud-vehicle-end-collaboration-based exercisable regional ODD platform device according to claim 1 is characterized in that a short-term OTA updating mode of the version management and issuing module is that non-ODD configuration files and intelligent driving software packages are integrated and issued in batches through OTA, and a long-term cloud platform updating mode is that non-ODD calibration files are independently stored and issued through a self-grinding cloud platform, and the vehicle-end is restarted to take effect after being downloaded automatically.

Description

Can exercise regional ODD platform device based on high in clouds-car end is cooperated Technical Field The invention relates to the technical field of intelligent driving, in particular to a movable area ODD platform device based on cloud-vehicle end cooperation. Background Under the background of rapid development of intelligent driving technology, the high-order auxiliary driving function has extremely high dependence on the accuracy and timeliness of a high-precision map. However, the actual road often changes frequently due to construction, renovation, etc., resulting in the disconnection of high-precision map information from the actual road conditions. The current map manufacturer is subject to the processes of data acquisition, auditing and release, the map updating period is more than three months, and during the period, the intelligent driving vehicle running on the old map is easy to have control errors, such as steering wheel steering, lane flushing and the like, so that serious safety risks are brought. The closest scheme in the prior art is an online freshness detection scheme based on a perception map. According to the scheme, the current road is subjected to online map building through the vehicle sensing module and is matched with the vehicle-mounted high-precision map in real time, and if the current road is inconsistent with the vehicle-mounted high-precision map, the driving function degradation is triggered. However, the scheme has obvious disadvantages of 1) limited coverage, detection blind areas, 2) passive update period, degradation only when problems occur, and incapability of early warning, 3) lack of cloud data closed-loop management, incapability of accumulating and multiplexing data, and 4) abrupt degradation, and influence on user experience. Therefore, it is necessary to provide a new exercisable area ODD platform device based on cloud-vehicle end collaboration to solve the above technical problems. Disclosure of Invention The technical problem to be solved by the invention is to provide the exercisable area ODD platform device based on cloud-vehicle end cooperation, which realizes the full-flow closed-loop management from acquisition, processing, updating to execution and feedback of non-ODD area data by constructing a three-layer cooperative architecture. In order to solve the technical problems, the cloud-vehicle-end-collaboration-based exercisable area ODD platform device provided by the invention comprises a cloud core layer, a vehicle-end execution layer and a tool supporting layer, wherein the cloud core layer, the vehicle-end execution layer and the tool supporting layer are operated cooperatively; the cloud core layer is used for storing, processing, version management and issuing of non-ODD region data, and comprises: The cloud database is used for storing structured data of a non-ODD region, wherein the structured data at least comprises a high-precision map version number, an associated lane ID, an ODD type, an abnormal starting point and an abnormal ending point; The data processing and updating module is used for executing the new adding, deleting and deduplication logic of the non-ODD area data; The version management and issuing module is used for issuing the non-ODD area data to the vehicle-end execution layer in a short-term OTA updating or long-term cloud platform updating mode; The vehicle end execution layer is integrated in the intelligent driving vehicle-mounted system and is used for data receiving, real-time calculation, mode switching and early warning prompt, and the vehicle end execution layer comprises: The data storage module is used for storing the non-ODD region data issued from the cloud core layer; the real-time sensing and calculating module is used for inquiring the data of the non-ODD area in real time and calculating the distance between the vehicle and the target non-ODD area according to the current position and the navigation path of the vehicle; the decision and execution module is used for executing multi-stage early warning prompt and driving mode switching according to the distance and the ODD type; The data returning module is used for returning the vehicle operation data to the cloud core layer when the preset conditions are met; The tool supporting layer provides operation and maintenance and verification tools for the cloud core layer and the vehicle end execution layer and at least comprises a problem screening tool, an ODD data manufacturing tool, a simulation verification tool and a visual operation and maintenance tool. Preferably, the duplication elimination logic of the data processing and updating module is configured to automatically discard the newly added data when the difference between coordinates of the abnormal start point and the abnormal end point of the newly added data and the coordinates recorded in the cloud database is less than or equal to 10 meters. Preferably, the execution logic of the decision and exec