An end-to-end system for data generation, map creation using the generated data, and localization to the created map is disclosed. Mapstreams—or streams of sensor data, perception outputs from deep neural networks (DNNs), and/or relative trajectory data—corresponding to any number of drives by any number of vehicles may be generated and uploaded to the cloud. The mapstreams may be used to generate map data—and ultimately a fused high definition (HD) map—that represents data generated over a plurality of drives. When localizing to the fused HD map, individual localization results may be generated based on comparisons of real-time data from a sensor modality to map data corresponding to the same sensor modality. This process may be repeated for any number of sensor modalities and the results may be fused together to determine a final fused localization result.

The systems may include dividing a digital map provided by a mapping system into a matrix having a plurality of cells; assigning a cell of the plurality of cells to encompass a geographic region of the digital map; calculating a number of sites of interest in the cell; creating a marker comprising a first count number representing the number of sites of interest in the cell; and sharing the marker with a browser for display on the digital map.

A map construction method and a related apparatus are provided. The method includes: obtaining, based on manual driving track data and/or an obstacle grid map, road information, intersection information, and lane information of a region through which a vehicle has traveled; obtaining road traffic direction information based on the manual driving track data and the road information, and obtaining lane traffic direction information based on the lane information and the road traffic direction information; obtaining intersection entry and exit point information based on the intersection information and the lane traffic direction information; and performing, based on the intersection entry and exit point information, an operation of generating a virtual topology center line to obtain an autonomous driving map of the region through which the vehicle has traveled, where the virtual topology center line is a traveling boundary line of the vehicle in an intersection region.

A computer-implemented method for controlling an excavation task by an autonomous excavation vehicle comprising a scanning device, the excavation task being described by a target map, the method comprising using an excavation vehicle control system for: a) according to data from the scanning device, maintaining a map representing current terrain; b) moving a sensor-equipped digging implement for executing an excavation operation; c) receiving data indicative of current terrain topography from the sensor; d) updating the maintained map according to the data indicative of current terrain topography; and e) calculating an excavation operation according to the difference between the maintained map and the target map.

Techniques described herein may be used to identify anomalies in a joined region of two datasets. For example, a computer system may determine an elevation anomaly data point in based on an elevation criterion. The computer system may determine a set of buffer points surrounding the elevation anomaly data points. The computer system may user the set of buffer points to reduce the number of data points for adjusting. The computer system may determine a reduction factor based on the buffer points. The computer system may then apply the reduction factor to points in the buffer points and other points.

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.

The present disclosure generally relates to a methods and systems for compensating for changes in the absolute position of locations with respect to the Earth's surface which occur over time due to crustal dynamics. The invention is particularly, although not exclusively, concerned with such compensation in the context of methods using digital map data, for example, methods of localization of a vehicle.

The disclosure provides a method and apparatus for constructing a navigation element in a map, a computer device and a readable storage medium. Various kinds of road information are recognized through semantic map data, roads are divided through lane lines, and road elements are added to the divided road blocks according to location information of the road elements so as to obtain a navigation map.

Aspects concern a method for correcting errors in map data comprising obtaining map data specifying travel possibilities between locations, generating a routing graph from the map data by assigning a vertex to each location and an edge from one vertex to another vertex if the map data specifies that it is possible to travel from the location to which the first vertex is assigned to the location to which the second vertex is assigned, identifying the largest strongly connected component of the routing graph, identifying one or more further strongly connected components of the routing graph, detecting errors in the map data by identifying travel possibilities that are not in the map data which lead to missing connections between the one or more further strongly connected components of the routing graph and the largest strongly connected component of the routing graph and amending the map data by the identified travel possibilities.

The present disclosure provides a device and method for generating lane information. The method includes obtaining information on a structure located around a vehicle and driving information of a surround vehicle based on information collected by a sensor and prestored navigation information; and generating geometric information on a travel lane of the vehicle in a road area based on the information on the structure and the driving information of the surround vehicle.