ADAS domain control method based on S32K series chip

Abstract

The invention discloses an ADAS domain control method based on an S32K series chip, and the method comprises the steps: building a communication protocol between an MCU and a vehicle computer vision processing unit VPU, transmitting radar data to the vehicle computer vision processing unit VPU through the communication protocol by the MCU, fusing the image recognition data with the radar data transmitted by the MCU by the vehicle computer vision processing unit VPU, and transmitting the fused target data back to the MCU for decision control by the vehicle computer vision processing unit VPU. According to the invention, the two systems are successfully fused into the same S32K-based embedded control platform, the functions of the two systems are realized by using the cost of a single system, and the cost is saved.

Description

technical field [0001] The invention relates to an ADAS domain control method, in particular to an ADAS domain control method based on S32K series chips, belonging to the technical field of automatic driving. Background technique [0002] The advanced driving assistance system uses various sensors or perception systems installed on the car to collect environmental data inside and outside the car at the first time, and perform technical processing such as identification, detection and tracking of static and dynamic objects. It is an active safety technology that enables the vehicle to detect possible dangers in the fastest time, and reminds the driver or even intervenes in vehicle control to avoid danger. The automatic parking system uses the perception system around the vehicle, such as ultrasonic radar or surround-view camera to identify the surrounding environment of the vehicle and dynamic and static objects, thereby extracting the coordinate information of the available ...

AI Technical Summary

Problems solved by technology

[0003] At present, the mainstream solution in the industry is that the advanced driver assistance system (forward) is controlled in real time by a single domain controller, while the automatic parking system is controlled by another separate controller. High, high cost for terminal customers to choose and install. The reason is that the hardware platforms and development resource systems of the two systems cannot be connected with each other. The advanced driver assistance system and the automatic parking system can be divided into three main functions: perception, decision-making and control. , the development of advanced driver assistance systems in the field of perception mostly relies on integrated vision suppliers, such as Mobileye, Horizon and other visual perception chip providers, and integrates the perception fusion function into another embedded control MCU, because The fusion algorithm requires high computing resources and occupies a lot of computing power of the current mainstream embedded MCU. As a result, after a single MCU platform completes the perception fusion calculation, the redundant computing power is only enough to support the final decision-making and control of the vehicle, while the automatic Although the ultrasonic radar information fusion and vehicle control required by the parking system do not require high computing power, they can no longer be realized on the same MCU. A new MCU platform is still needed to realize it, resulting in two hardware platforms and a large number of underlying development. work; the cost is high, and the development cost increases, and the OEM needs to invest more unit cost, R&D cost and additional engineering resources to develop the two systems

Images