A system, a method and instructions embodied on a non-transitory computer-readable storage medium that solve a 3D point-in-polygon (PIP) problem is presented. This system projects polygons that comprise a set of polyhedra onto projected polygons in a reference plane. Next, the system projects a data point onto the reference plane, and performs a 2D PIP operation in the reference plane to determine which projected polygons the projected data point falls into. For each projected polygon the projected data point falls into, the system performs a 3D crossing number operation by counting intersections between a ray projected from the corresponding data point in a direction orthogonal to the reference plane and polyhedral faces corresponding to projected polygons, to identify polyhedra the data point falls into. The system then generates a visual representation of the set of polyhedra, wherein each polyhedron is affected by data points that fall into it.

Embodiments relate to methods for efficiently encoding sensor data captured by an autonomous vehicle and building a high definition map using the encoded sensor data. The sensor data can be LiDAR data which is expressed as multiple image representations. Image representations that include important LiDAR data undergo a lossless compression while image representations that include LiDAR data that is more error-tolerant undergo a lossy compression. Therefore, the compressed sensor data can be transmitted to an online system for building a high definition map. When building a high definition map, entities, such as road signs and road lines, are constructed such that when encoded and compressed, the high definition map consumes less storage space. The positions of entities are expressed in relation to a reference centerline in the high definition map. Therefore, each position of an entity can be expressed in fewer numerical digits in comparison to conventional methods.

An image localization system is described. A server stores a localization model that associates location and orientation data with each picture among a plurality of pictures. The server receives a query for a location and an orientation of a device. The query includes a first picture. A second picture that matches the first picture is identified. The server determines the location and the orientation of the device based on the location and orientation data corresponding to the second picture. The location and orientation data indicates position, orientation, three-dimensional geometry, gyroscope measurement, and accelerometer measurement.

The disclosure relates to using ambient lighting conditions with passenger and goods pickups and drop offs with autonomous vehicles. For instance, a map of ambient lighting conditions for stopping locations may be generated by receiving ambient lighting condition data for predetermined stopping locations and arranging this data into a plurality of buckets based on time and one of the stopping locations. A vehicle may then be controlled in an autonomous driving mode in order to stop for a passenger by both observing ambient lighting conditions for different stopping locations and, in some instances, also using the map.

A mobile robot includes a receiving unit that receives designation of a destination region including a destination, a moving unit that moves toward the destination region, and a seeking unit that seeks a client after movement toward the destination region starts. The moving unit moves toward the sought client.

A method includes a mobile device receiving data from a first beacon in a plurality of beacons, wherein the data indicates a plurality signals strengths of signals received by the first beacon from the other beacons in the plurality of beacons. The mobile device uses multivariable regression based on the data received from the first beacon and signal strengths of signals received by the mobile device from the other beacons in the plurality of beacons to determine distances from the mobile device to the beacons. The mobile device determines a location of the mobile device based on the determined distances.

Aspects concern a distributed computing system for generating vector tiles of a selected map area including a memory unit configured to store map data and a task database, the map data including a representation of the selected map area with a first resolution and a first detail level and with a second resolution higher than the first resolution and a second detail level; two or more processing units, each of the two or more processing units configured to select a task included in the task database, to execute the selected task, and to provide data generated by the selected task to the memory unit for storage; wherein one of the two or more processing units is further configured to schedule the generation of vector tiles by determining tasks using a specific predefined directed acyclic task graph and to provide the determined task to the task database.

A-implemented method initializes a localization system for a vehicle, the localization system using map data comprising a plurality of maps for localizing the vehicle in an area The method includes associating map related information with a last available position of the vehicle which was used by the localization system before shutdown of the localization system, wherein the map related information comprises a unique identifier for a first map of the plurality of maps which was used for localizing the vehicle at the last available position, at start-up of the localization system with respect to the last available position, loading a map from the plurality of maps for localizing the vehicle at the last available position and verifying if the loaded map is the first map by using the unique identifier, and when it is verified that the first map is loaded, initializing the localization system with the last available position in the first map.

A high-definition map building method includes determining a key node describing information about a key position of a lane attribute change, determining a key node layer based on a position of the key node and an attribute of the key node, and determining a high-definition map based on a navigation map and the key node layer. The navigation map provides road-level navigation information, and the high-definition map provides lane-level navigation information.

Provided is a map storage device that generates a map by automatically determining and storing a necessary map without a user's operation. The map storage device includes: a map creation unit that creates map data of a route traveled by a vehicle; a map data temporary storage unit that temporarily stores the map data; a map data storage unit that non-temporarily stores a part of the map data temporarily stored in the map data temporary storage unit; a behavior history information storage unit that stores behavior history information that is a behavior history of the vehicle; and a vehicle behavior determination unit that determines whether or not to store, in the map data storage unit, the map data temporarily stored in the map data temporary storage unit, based on the behavior history information.