Abstract: This application provides example terminal device localization methods and related devices. One example method includes obtaining, from a vector map, a first map point matching a first feature point in a current image frame shot by a terminal device. A second map point matching a second feature point in another image frame before the current image frame is obtained from the vector map. A pose in which the terminal device shoots the current image frame is adjusted based on a target function to obtain, as a localization result of the terminal device, a pose in which the terminal device shoots the current image frame and that is obtained after current adjustment. The target function includes a first matching error between the first feature point and the first map point and a second matching error between the second feature point and the second map point.

Abstract: A mobile device positioning method and a mobile device positioning apparatus (900) are disclosed. Perception vector elements (301 and 302) are associated with projection vector elements (303 and 304) based on horizontal positions of the perception vector elements (301 and 302) extracted from an image and horizontal positions of the projection vector elements (303 and 304) obtained through projection of vector elements in a map. In this way, when the horizontal positions of the associated perception vector elements (301 and 302) and the horizontal positions of the associated projection vector elements (303 and 304) are separately arranged in a same order, horizontal position distribution of all the arranged perception vector elements (301 and 302) is consistent with that of corresponding projection vector elements (303 and 304) in the arranged projection vector elements (303 and 304), and pose estimation is performed on a mobile device based on an association result.

Abstract: This application relates to the field of artificial intelligence. The technology includes: obtaining data of each three-dimensional point in a first three-dimensional point set, where the data of each three-dimensional point is collected when an intelligent device is in a first position, and the first three-dimensional point set includes a plurality of three-dimensional points located in a first coordinate system; determining a first three-dimensional point subset from the first three-dimensional point set, where the first three-dimensional point subset forms a first plane; calculating a plurality of first distances, which is a distance between a first three-dimensional point and the first position; determining a target first three-dimensional point from the first three-dimensional point subset based on the plurality of first distances; and locating the intelligent device by using the target first three-dimensional point as a constraint to obtain a pose of the first position.

Abstract: One example method includes obtaining a state vector set, where the state vector set includes at least one of a pose vector of an intelligent device or a landmark characteristic vector of an environment in which the intelligent device is located, determining a first linearized constraint equation, determining a first state vector, marginalizing the first state vector in the first linearized constraint equation to obtain a second linearized constraint equation, determining an objective function based on the second linearized constraint equation, and optimizing the at least one of the at least one pose vector or the at least one landmark characteristic vector based on the objective function to output a track of the intelligent device and a map of the environment in which the intelligent device is located.

Abstract: A storage space management method and apparatus, where the method and apparatus are applied to a non-volatile memory (NVM). In a feature set that includes M image features of M idle blocks in storage space of the NVM, an idle block whose image feature is highly similar to an image feature of data to be written into the NVM is determined such that the data is written into the idle block. In this way, wear and energy consumption problems are considered during storage space allocation, and a write operation of an idle block in storage space of an NVM can consume less energy, thereby extending a life span of the NVM and reducing write operation energy consumption.

Abstract: A storage space management method and apparatus, where the method and apparatus are applied to a non-volatile memory (NVM). In a feature set that includes M image features of M idle blocks in storage space of the NVM, an idle block whose image feature is highly similar to an image feature of data to be written into the NVM is determined such that the data is written into the idle block. In this way, wear and energy consumption problems are considered during storage space allocation, and a write operation of an idle block in storage space of an NVM can consume less energy, thereby extending a life span of the NVM and reducing write operation energy consumption.