High definition mapping
In various examples, a high definition (HD) map is provided that includes a segmented data structure that allows for selective access to desired road segments and corresponding layers of map data. For example, the HD map may be segmented into a series of tiles that may correspond to a geographic region, and each of the tiles may include any number of road segments corresponding to portions of the geographic region. Each road segment may include a corresponding set of layers—which may include driving layers for use by the ego-machine and/or training layers for generating ground truth data—from the HD map that are associated with the road segment alone. As such, when traversing the environment, an ego-machine may determine one or more road segments within a tile corresponding to a current location, and may selectively download one or more layers for each of the one or more road segments.
1 . At least one processor comprising:
one or more circuits to:
access, based at least on a current location of an ego-machine, first localization data from a first road segment of road segments representing respective map regions aligned to one or more roads of a high definition (HD) map, the HD map including a graph having nodes representing the road segments,
wherein the first road segment stores the first localization data corresponding to a first region of the respective map regions and a second road segment of the road segments stores second localization data corresponding to a second region of the respective map regions;
perform one or more localization operations using the first localization data from the first road segment to determine first localization results of the first road segment, the first localization results including a location of the ego-machine within a first local coordinate system of the first road segment;
identify an edge in the graph based at least on the edge representing a road-level connection between the first road segment and the second road segment;
identify, using the edge, a spatial transformation between the first local coordinate system of the first road segment and a second local coordinate system of the second road segment;
translate, when the ego-machine traverses into the second road segment from the first road segment and using the spatial transformation, the first localization results from the first local coordinate system to second localization results of the second road segment in the second local coordinate system, the second localization results including a translated location of the ego-machine; and
perform one or more of autonomously controlling at least one of steering or speed of the ego-machine using the second localization data and the translated location in the second local coordinate system.
2 . The at least one processor of claim 1 , wherein the one or more circuits are further to:
determine, based at least on the current location, a tile from a tile listing, the tile corresponding to the current location of the ego-machine and including the first road segment and the second road segment.
3 . The at least one processor of claim 1 , wherein the first road segment includes a first layer of a first sensor modality localization layer type and a second layer of a second sensor modality localization layer type, and the second road segment includes a third layer of the first sensor modality localization layer type and a fourth layer of the second sensor modality localization layer type.
4 . The at least one processor of claim 3 , wherein the first sensor modality localization layer type comprises camera data and the second sensor modality localization layer type comprises LiDAR data.
5 . The at least one processor of claim 1 , wherein the first localization data encodes poles representing environmental objects as respective three-dimensional straight lines in the first local coordinate system, and the one or more localization operations include determining the translated location of the ego-machine using the respective three-dimensional straight lines.
6 . The at least one processor of claim 1 , wherein the spatial transformation comprises a coordinate transformation encoded by a quaternion corresponding to a three-dimensional orientation of the ego-machine and a three-dimensional translation vector corresponding to the location.
7 . The at least one processor of claim 1 , wherein the HD map includes the graph having the nodes corresponding to the road segments and edges corresponding to road-level connections between the road segments, and the respective map regions are respective circular regions that are partially overlapping with one another.
8 . The at least one processor of claim 1 , wherein each road segment of the road segments includes a first universally unique identifier (UUID) and each layer type corresponding to each road segment includes a second UUID, and the access is facilitated using the first UUID and the second UUID.
9 . The at least one processor of claim 1 , wherein the first region partially overlaps with the second region.
10 . The at least one processor of claim 1 , wherein respective origins of the road segments are distributed along and within the one or more roads.
11 . The at least one processor of claim 1 , wherein the first road segment includes one or more of:
a junctions layer that encodes jurisdictional rules or regulations;
a lane channel layer that encodes contiguous boundaries corresponding to one or more lanes on a driving surface;
a road boundary layer that encodes a height channel corresponding to a height of a road boundary; or
a divider layer that encodes a height channel corresponding to a height of a divider.
12 . The at least one processor of claim 1 , wherein the first road segment includes driving layers and a learning layer including data used to generate ground truth data for training one or more deep neural networks (DNNs).
13 . The at least one processor of claim 1 , wherein the first localization data includes a first sensor modality localization layer type and a second sensor modality localization layer type that are selectively accessed to perform the one or more localization operations based at least on the ego-machine having one or more corresponding sensor types.
14 . The at least one processor of claim 13 , wherein the second sensor modality localization layer type is accessed based at least on a dependency of the first sensor modality localization layer type on the second sensor modality localization layer type.
15 . The at least one processor of claim 1 , wherein the at least one processor is comprised in at least one of:
a control system for an autonomous or semi-autonomous machine;
a perception system for an autonomous or semi-autonomous machine;
a system for performing simulation operations;
a system for collaborative content creation;
a system for performing deep learning operations;
a system implemented using an edge device;
a system implemented using a robot;
a system for generating synthetic data;
a system incorporating one or more virtual machines (VMs);
a system implemented at least partially in a data center; or
a system implemented at least partially using cloud computing resources.
16 . A method comprising:
accessing, based at least on a current location of an ego-machine, first localization data from a first road segment of road segments representing respective map regions aligned to one or more roads of a high definition (HD) map, the HD map including a graph having nodes representing the road segments,
wherein the first road segment stores the first localization data corresponding to a first region of the respective map regions and a second road segment of the road segments stores second localization data corresponding to a second region of the respective map regions;
performing one or more localization operations using the first localization data from the first road segment to determine first localization results of the first road segment, the first localization results including a location of the ego-machine within a first local coordinate system of the first road segment;
identifying an edge in the graph based at least on the edge representing a road-level connection between the first road segment and the second road segment;
identifying, using the edge, a spatial transformation between the first local coordinate system of the first road segment and a second local coordinate system of the second road segment;
translating, when the ego-machine traverses into the second road segment from the first road segment and using the spatial transformation, the first localization results from the first local coordinate system to second localization results of the second road segment in the second local coordinate system, the second localization results including a translated location of the ego-machine; and
performing one or more of autonomously controlling at least one of steering or speed of the ego-machine using the second localization data and the translated location in the second local coordinate system.
17 . The method of claim 16 , wherein the method further includes:
determining, based at least on the current location, a tile from a tile listing, the tile corresponding to the current location of the ego-machine and including the first road segment and the second road segment.
18 . The method of claim 16 , wherein the first road segment includes a first layer of a first sensor modality localization layer type and a second layer of a second sensor modality localization layer type, and the second road segment includes a third layer of the first sensor modality localization layer type and a fourth layer of the second sensor modality localization layer type.
19 . The method of claim 18 , wherein the first sensor modality localization layer type comprises camera data and the second sensor modality localization layer type comprises LiDAR data.
20 . The method of claim 16 , wherein the first road segment includes at least one of a lane channel layer, a lane marking layer, a junctions layer, a lane planning layer, a path planning layer, a localization layer, a RADAR localization layer, a LiDAR localization layer, a camera localization layer, or a road boundary or divider height layer.
21 . The method of claim 16 , wherein the method further includes performing at least one of one or more planning or actuation operations using the road segments of the HD map and the ego-machine.
22 . The method of claim 16 , wherein the road segments are distributed across a plurality of tiles of the HD map.
23 . The method of claim 16 , wherein each road segment of the road segments includes a first universally unique identifier (UUID) and each layer type corresponding to each road segment includes a second UUID, and the access to the first localization data is facilitated using the first UUID and the second UUID.
24 . A system comprising:
one or more processors to perform operations including:
accessing, based at least on a current location of an ego-machine, first localization data from a first road segment of road segments representing respective map regions aligned to one or more roads of a high definition (HD) map, the HD map including a graph having nodes representing the road segments,
wherein the first road segment stores the first localization data corresponding to a first region of the respective map regions and a second road segment of the road segments stores second localization data corresponding to a second region of the respective map regions;
performing one or more localization operations using the first localization data from the first road segment to determine first localization results of the first road segment, the first localization results including a location of the ego-machine within a first local coordinate system of the first road segment;
identifying an edge in the graph based at least on the edge representing a road-level connection between the first road segment and the second road segment;
identifying, using the edge, a spatial transformation between the first local coordinate system of the first road segment and a second local coordinate system of the second road segment;
translating, when the ego-machine traverses into the second road segment from the first road segment and using the spatial transformation, the first localization results from the first local coordinate system to second localization results of the second road segment in the second local coordinate system, the second localization results including a translated location of the ego-machine; and
performing one or more of autonomously controlling at least one of steering or speed of the ego-machine using the second localization data and the translated location in the second local coordinate system.
25 . The system of claim 24 , wherein the operations further include:
determining, based at least on the current location, a tile from a tile listing, the tile corresponding to the current location of the ego-machine and including the first road segment and the second road segment.
26 . The system of claim 24 , wherein the first road segment includes a first layer of a first sensor modality localization layer type and a second layer of a second sensor modality localization layer type, and the second road segment includes a third layer of the first sensor modality localization layer type and a fourth layer of the second sensor modality localization layer type.
27 . The system of claim 26 , wherein the first sensor modality localization layer type comprises camera data and the second sensor modality localization layer type comprises LiDAR data.
28 . The system of claim 24 , wherein the first road segment includes at least one of a lane channel layer, a lane marking layer, a junctions layer, a lane planning layer, a path planning layer, a localization layer, a RADAR localization layer, a LiDAR localization layer, a camera localization layer, or a road boundary or divider height layer.
29 . The system of claim 24 , wherein the operations further include performing at least one of one or more planning or actuation operations using the road segments of the HD map and the ego-machine.
30 . The system of claim 24 , wherein the system is comprised in at least one of:
a control system for an autonomous or semi-autonomous machine;
a perception system for an autonomous or semi-autonomous machine;
a system for performing simulation operations;
a system for collaborative content creation;
a system for performing deep learning operations;
a system implemented using an edge device;
a system implemented using a robot;
a system for generating synthetic data;
a system incorporating one or more virtual machines (VMs);
a system implemented at least partially in a data center; or
a system implemented at least partially using cloud computing resources.