Abstract: A vehicle control method includes obtaining sensor status information and autonomous driving function information, where the sensor status information includes information indicating that one or more sensors are in a failed state. The autonomous driving function information indicates a currently running autonomous driving function. The autonomous driving function includes a plurality of running conditions, and implementation of each running condition is related to one or more sub-functions. A driving policy is implemented based on the impact of a failure of the one or more sensors on the running condition and the currently running autonomous driving function.

Abstract: A self-driving safety evaluation method, including determining RB based on a risk value of a vehicle in a shadow driving mode in a first measurement unit, where RB is a risk value of the vehicle in the shadow driving mode in a plurality of measurement units, and determining RC based on a risk value of the vehicle in a self-driving mode based on a preset route in the first measurement unit, where RC is a risk value of the vehicle in the self-driving mode based on the preset route in the measurement units, where RB and RC are used to determine whether safety of the vehicle in the self-driving mode meets a requirement.

Abstract: A method and a computing device for object detection and tracking from a video input are described. The method and the computing device may be used to, for example, track objects of interest, such as lane markings, in traffic. A plurality of frames corresponding to a video may be analyzed in a spatiotemporal domain by a neural network. The neural network may be trained using data synthesized in the spatiotemporal domain.

Abstract: This application discloses a method for determining a lane line recognition abnormal event, and a lane line recognition apparatus and system. One example method includes: determining an updated lane line recognition confidence based on a posteriori lane line recognition result; constructing a lane line true value based on the updated lane line recognition confidence; and determining a lane line recognition abnormal event based on the lane line recognition result and the lane line true value.

Abstract: A self-driving safety evaluation method, including determining RB based on a risk value of a vehicle in a shadow driving mode in a first measurement unit, where RB is a risk value of the vehicle in the shadow driving mode in a plurality of measurement units, and determining RC based on a risk value of the vehicle in a self-driving mode based on a preset route in the first measurement unit, where RC is a risk value of the vehicle in the self-driving mode based on the preset route in the measurement units, where RB and RC are used to determine whether safety of the vehicle in the self-driving mode meets a requirement.