Chassis domain controller for intelligent vehicle, vehicle control method and vehicle

Abstract

The invention provides a chassis domain controller for an intelligent vehicle, a vehicle control method and the vehicle. The chassis domain controller comprises a signal processing unit, a logic verification unit and a model solving unit which are connected with one another; the signal processing unit is connected with a plurality of sensors of the vehicle and used for obtaining original signals detected by the sensors, the original signals are processed to obtain a driving signal reflecting a vehicle state or a road environment, and the driving signal is sent to the logic verification unit and the model solving unit; the model solving unit is used for receiving the driving signal, carrying out model solving based on the driving signal and determining a decision result; and the logic verification unit is connected with the plurality of sensors and is used for determining a verification signal according to the original signals, comparing the verification signal with the driving signal so as to realize the verification of the driving signal, performing verification calculation according to the verification signal, determining a verification result, and comparing the verification result with the decision result so as to realize the verification of the decision result.

Description

technical field [0001] The present application relates to the technical field of automobiles, in particular, to a chassis domain controller for intelligent automobiles, a vehicle control method, and the vehicle. Background technique [0002] With the continuous improvement of automobile technology, the use of advanced sensors and intelligent algorithms in automobiles is increasing. However, these advanced algorithms and advanced sensors often rely on the actuators of the automobile chassis. At the same time, these algorithms and sensors themselves may have their own function failure or the possibility of insufficient expected functions. [0003] There are two directions for the evolution of the controller on the vehicle. One is that the number of ECUs (Electronic Control Units) in the chassis continues to increase, which leads to the dispersion and inconsistency of the functional sub-modules of the chassis and is susceptible to false triggers due to information changes in ot...

AI Technical Summary

Problems solved by technology

However, these advanced algorithms and advanced sensors often rely on the actuators of the car chassis. At the same time, these algorithms and sensors themselves may have their own function failure or the possibility of insufficient expected functions.

[0003] There are two directions for the evolution of the controller on the vehicle. One is that the number of ECUs (Electronic Control Units) in the chassis continues to increase, which leads to the dispersion and inconsistency of the functional sub-modules of the chassis and is susceptible to false triggers due to information changes in other modules. Scattered dimensionality reduction models cannot cope with the existing complex vehicle driving conditions and the uncertainty of the road environment. Smart cars face uncontrollable factors when running under such conditions, and it is difficult to deal with emergency and dangerous situations

The other is at the sensor level, a heterogeneous computing mode of a single industrial computer plus a GPU (Graphics Processing Unit, display chip) is often used to fuse sensor information such as lidar, millimeter-wave radar, and camera, which is difficult to make full use of The sensor signal makes the determined driving decision too conservative, unable to make full use of the actual performance of the vehicle and the road surface

In addition, it is difficult for the existing intelligent driving to identify the state of low road adhesion and idealize the driving state of the vehicle. At the same time, the robustness of software and hardware is not strong and it is easy to cause potential safety hazards.

Images