Lane group modification and generation
Techniques for changing driving lanes due to construction zones are described herein. A vehicle may receive sensor data and determine, based on the sensor data, that the vehicle is approaching a construction zone. The vehicle may determine that the construction zone is blocking the driving lane occupied by the vehicle. In this case, the vehicle can generate a driving corridor that represents a driving lane within which the vehicle can navigate. The vehicle can use the driving corridor to generate a first lane group that is specific to navigating around the construction zone. Further, the vehicle can generate a second lane group based on map data. The vehicle can generate an updated lane group by combining the first lane group with the second lane group. The vehicle can be controlled based on the updated lane group.
1 . A system comprising:
one or more processors; and
one or more non-transitory computer-readable media storing computer-executable instructions that, when executed, cause the system to perform operations comprising:
receiving, from a sensor associated with an autonomous vehicle, sensor data representative of an environment;
detecting, based at least in part on the sensor data, that a driving lane in which the autonomous vehicle is currently traversing is associated with a portion that is restricted for driving;
receiving, based at least in part on the driving lane being restricted, a driving corridor that represents a region of the environment available for the autonomous vehicle to traverse to avoid the portion that is restricted for driving;
determining a path associated with the driving corridor;
determining, based at least in part on map data, a group of lane segments that is different than the path;
determining, based at least in part on adding the path to the group of lane segments, an updated group of lane segments; and
controlling movement of the autonomous vehicle based at least in part on the updated group of lane segments.
2 . The system of claim 1 , wherein the portion is a restricted area, wherein determining that the driving lane is restricted is based at least in part on:
generating, based at least in part on the sensor data and a machine-learned model, a zone contour that outlines the restricted area; and
determining, based at least in part on the zone contour and the map data, that the driving lane is restricted.
3 . The system of claim 2 , wherein determining that the driving lane is restricted is based at least in part on:
determining, based at least in part on the map data, a distance from a side of the zone contour that outlines the restricted area to a lane marker defining a border of the driving lane;
determining that a width of the autonomous vehicle meets or exceeds the distance; and
determining, based at least in part on the width meeting or exceeding the distance, that the driving lane is restricted.
4 . The system of claim 1 , wherein the portion is a restricted area, wherein determining the driving corridor is based at least in part on:
generating a polygon representing the region of the environment available for the autonomous vehicle to traverse to avoid the restricted area; and
generating, as the driving corridor and based at least in part on a width of the autonomous vehicle, a modified polygon that includes a different width than the polygon.
5 . The system of claim 1 , wherein the controlling the autonomous vehicle is based at least in part on:
determining that the updated group of lane segments includes a merging scenario between the driving lane and a laterally spaced driving lane;
detecting an object in the laterally spaced driving lane;
determining, based at least in part on the updated group of lane segments, a first type of lane segment associated with the laterally spaced driving lane, the first type of lane segment being associated with a first weight;
determining a second weight of a second type of the path associated with the driving corridor, wherein the second weight is higher than the first weight;
determining, based at least in part on the first weight and the second weight, a right-of-way score associated with the object; and
causing, based at least in part on the right-of-way score, the autonomous vehicle to enter a merged driving lane in front of the object.
6 . The system of claim 1 , wherein the group of lane segments comprises one or more connected lane segments that lack lane changes, and wherein the path is associated with multiple driving lanes.
7 . One or more non-transitory computer-readable media storing instructions executable by one or more processors, wherein the instructions, when executed, cause a system to perform operations comprising:
receiving sensor data representative of an environment;
detecting, based at least in part on the sensor data, that an autonomous vehicle is approaching a restricted area;
receiving, based at least in part on the restricted area, a driving corridor that represents a region of the environment available for the autonomous vehicle to traverse to avoid the restricted area;
determining a path associated with the driving corridor;
determining, based at least in part on map data, a group of lane segments that is different than the path;
determining, based at least in part on adding the path to the group of lane segments, an updated group of lane segments; and
controlling movement of the autonomous vehicle based at least in part on the updated group of lane segments.
8 . The one or more non-transitory computer-readable media of claim 7 , wherein receiving the driving corridor is based at least in part on:
determining that a driving lane occupied by the autonomous vehicle is restricted by the restricted area.
9 . The one or more non-transitory computer-readable media of claim 8 , wherein determining that the driving lane is restricted is based at least in part on:
generating, based at least in part on a machine-learned model, a zone contour that outlines the restricted area; and
determining, based at least in part on the zone contour and the map data, that the driving lane is restricted.
10 . The one or more non-transitory computer-readable media of claim 9 , wherein determining that the driving lane is restricted is based at least in part on:
determining, based at least in part on the map data, a distance from a side of the zone contour that outlines the restricted area to a lane marker defining a border of the driving lane;
determining that a width of the autonomous vehicle meets or exceeds the distance; and
determining, based at least in part on the width meeting or exceeding the distance, that the driving lane is restricted.
11 . The one or more non-transitory computer-readable media of claim 7 , wherein determining the driving corridor is based at least in part on:
generating a polygon representing the region of the environment available for the autonomous vehicle to traverse to avoid the restricted area; and
generating, as the driving corridor and based at least in part on a width of the autonomous vehicle, a modified polygon that includes a different width than the polygon.
12 . The one or more non-transitory computer-readable media of claim 7 , wherein the group of lane segments comprises one or more connected lane segments that lack lane changes, and wherein the path is associated with multiple driving lanes.
13 . The one or more non-transitory computer-readable media of claim 7 , wherein the controlling the autonomous vehicle is based at least in part on:
determining that the updated group of lane segments includes a merging scenario between a driving lane of the autonomous vehicle and a laterally spaced driving lane;
detecting an object in the laterally spaced driving lane;
determining, based at least in part on the updated group of lane segments, a first type of lane segment associated with the laterally spaced driving lane, the first type of lane segment being associated with a first weight;
determining a second weight of a second type of path associated with the driving corridor, wherein the second weight is higher than the first weight;
determining, based at least in part on the first weight and the second weight, a right-of-way score associated with the object; and
causing, based at least in part on the right-of-way score, the autonomous vehicle to enter a merged driving lane in front of the object.
14 . A method comprising:
receiving sensor data representative of an environment;
detecting, based at least in part on the sensor data, that an autonomous vehicle is approaching a restricted area;
receiving, based at least in part on the restricted area, a driving corridor that represents a region of the environment available for the autonomous vehicle to traverse to avoid the restricted area;
determining a path associated with the driving corridor;
determining, based at least in part on map data, a group of lane segments that is different than the path;
determining, based at least in part on adding the path to the group of lane segments, an updated group of lane segments; and
controlling movement of the autonomous vehicle based at least in part on the updated group of lane segments.
15 . The method of claim 14 , wherein receiving the driving corridor is based at least in part on:
determining that a driving lane occupied by the autonomous vehicle is restricted by the restricted area.
16 . The method of claim 15 , wherein determining that the driving lane is restricted is based at least in part on:
generating, based at least in part on a machine-learned model, a zone contour that outlines the restricted area; and
determining, based at least in part on the zone contour and the map data, that the driving lane is restricted.
17 . The method of claim 16 , wherein determining that the driving lane is restricted is based at least in part on:
determining, based at least in part on the map data, a distance from a side of the zone contour that outlines the restricted area to a lane marker defining a border of the driving lane;
determining that a width of the autonomous vehicle meets or exceeds the distance; and
determining, based at least in part on the width meeting or exceeding the distance, that the driving lane is restricted.
18 . The method of claim 14 , wherein determining the driving corridor is based at least in part on:
generating a polygon representing the region of the environment available for the autonomous vehicle to traverse to avoid the restricted area; and
generating, as the driving corridor and based at least in part on a width of the autonomous vehicle, a modified polygon that includes a different width than the polygon.
19 . The method of claim 14 , wherein the group of lane segments comprises one or more connected lane segments that lack lane changes, and wherein the path is associated with multiple driving lanes.
20 . The method of claim 19 , wherein the controlling the autonomous vehicle is based at least in part on:
determining that the updated group of lane segments includes a merging scenario between a driving lane of the autonomous vehicle and a laterally spaced driving lane;
detecting an object in the laterally spaced driving lane;
determining, based at least in part on the updated group of lane segments, a first type of lane segment associated with the laterally spaced driving lane, the first type of lane segment being associated with a first weight;
determining a second weight of a second type of path associated with the driving corridor, wherein the second weight is higher than the first weight;
determining, based at least in part on the first weight and the second weight, a right-of-way score associated with the object; and
causing, based at least in part on the right-of-way score, the autonomous vehicle to enter a merged driving lane in front of the object.