A method, apparatus and computer program product are provided for learning to generate maps from raw geospatial observations from sensors traveling within an environment. Methods may include: receiving a plurality of sequences of geospatial observations from discrete trajectories; aligning the trajectories to generate aligned geospatial observations; concatenating the aligned geospatial observations; processing the concatenated, aligned geospatial observations using one or more Set Transformers; generating, from the at least one Set Transformer, map geometries including objects from the geospatial observations; and providing at least one of navigational assistance or at least semi-autonomous vehicle control based on the map geometries. According to some embodiments, aligning the trajectories includes applying a geospatial offset for one or more of the trajectories.

According to the present disclosure, a map matching apparatus includes a map information acquisition unit configured to acquire map information expressed by a plurality of links, a movement information acquisition unit configured to acquire movement information of a moving body on a predetermined route, and a matching unit configured to specify a link string, which is a string of the links corresponding to the route, based on the map information and the movement information, in which the matching unit is configured to generate a physical network, which is a road network, based on the connection relationship and the map information, generate a hierarchical logical network in which the physical network is duplicated in a plurality of layers and configured in a hierarchical manner, and specify the link string indicating a minimized cost in the hierarchical logical network.

Embodiments are disclosed for a feature-based simultaneous localization and mapping (SLAM) system and method that generates radio maps for environments that are not accessible for surveying. More accurate radio maps are generated for an unsurveyed environment by determining a best estimate of a mobile device state from harvested traced data that maximizes a posterior probability of the mobile device state given measurements, landmarks and loop constraints.

Techniques and systems are described for generating and using a sound localization model. A described technique includes obtaining for a building a sound sensor map indicating locations of first and second sound sensor devices in respective first and second rooms of the building; causing an autonomous device to navigate to the first room and to emit, during a time window, sound patterns at one or more frequencies within the first room; receiving sound data including first and second sound data respectively from the first and second sound sensor devices that are observed during the time window; and generating and storing a sound localization model based on the sound sensor map, autonomous device location information, and the received sound data, the model being configured to compensate for how sounds travels among rooms in at least a portion of the building such that an origin room of a sound source is identifiable.

An approach is provided for pattern-based map matching of a probe trace to a digital map. The approach involves querying the digital map for a set of road links within a threshold distance of a probe point. The approach also involves determining a match starting point for each road link of a set of road links. The approach further involves selecting a sampled probe point from the probe trace. The approach also involves generating one or more patterns for said each road link based on the match starting point, the sampled probe point, the topology polyline, or a combination thereof. The approach further involves selecting a matched road link from among the set of road links based on the one or more patterns. The probe point is then snapped to the matched road link.

A server includes a feature point position estimation section that estimates positions of feature points of a plurality of respective input maps if a position correction section has failed in position correction for an input map, an accumulated data generation section that accumulates the positions of the feature points of the plurality of respective input maps to generate accumulated data, a convergence determination section that determines whether the positions of the feature points have converged based on the accumulated data, a difference data generation section that generates, if it is determined that the positions of the feature points have converged, difference data from the positions of the feature points that have converged, and a difference data reflection section that reflects the difference data in the reference map to update the reference map.

The present disclosure provides a method and apparatus for automatically producing map data. The method includes: performing track rectification on crowdsourcing tracks based on corresponding standard tracks, and locating each map element included, based on depth information of track point images included in the rectified crowdsourcing tracks; comparing a latest map element obtained based on the rectified crowdsourcing tracks locating and an old map element at a corresponding locating position using a pre-built entity semantic map; determining, in response to a change in the latest map element compared to the old map element, a target processing method according to a processing standard of a changed map element pre-abstracted from a map element update specification; and processing the latest map element according to the target processing method to obtain a processed latest map.

Techniques for determining distortion in a map caused by measurement errors are discussed herein. For example, such techniques may include implementing a model to estimate map distortion between the map frame and the inertial frame. Data such as sensor data, map data, and vehicle state data may be input into the model. A map distortion value output from the model may be used to compensate vehicle operations in a local region by approximating the distortion as linearly varying about the region. A vehicle, such as an autonomous vehicle, can be controlled to traverse an environment based on the trajectory.

Methods and systems for distributed detection of road conditions and damage are described. In one embodiment, a method for distributed detection of road conditions and damage is provided. The method includes receiving, from one or more mobile devices, a plurality of reports of anomalies associated with roads in a geographic area. The method also includes storing the received reports of anomalies in a database and comparing each report of an anomaly to stored reports of previous anomalies in the database. The method further includes determining whether each report of an anomaly indicates road damage or a temporary problem. The method includes generating a prioritized list of locations of anomalies associated with one or more roads that have been determined to have road damage that needs maintenance and/or repair.

A computer-implemented method may include monitoring an age of a tile of a map, where the map includes multiple tiles including the tile. The method may also include, based on the age exceeding a threshold age, determining that the tile of the map is to be updated, and receiving a location indicator from a vehicle. The method may additionally include transmitting an update message to a vehicle traversing a track within the tile as indicated by the location indicator, where the update message includes instructions to cause the vehicle to gather and submit sensor data to a computing system. The method may also include receiving the sensor data from the vehicle, and updating the tile of the map based on the received sensor data.