Collision mitigation static occupancy grid
A system and method for avoiding or mitigating collisions by an autonomous vehicle is provided. Raw sensor data for the vehicle's environment is received, such as through a perception system of the vehicle. A trajectory is also received, the trajectory having a width corresponding to a distance that is at least as wide as the vehicle. The trajectory and raw sensor data are projected onto a grid, such as a static occupancy grid, having a plurality of cells. It is determined whether any cell that overlaps with the trajectory includes the raw sensor data. It is further determined, based on the raw sensor data, whether an obstacle is present along the trajectory. When no obstacles are determined to be present along the trajectory, the trajectory may be verified. However, where obstacles are determined to be present, the vehicle may take an action to avoid collision.
1. A method comprising:
receiving, by one or more processors, sensor data from a perception system of a vehicle, the sensor data indicating whether there are objects in an external environment of the vehicle;
projecting, by one or more processors, a trajectory of the vehicle onto a grid having a plurality of cells;
evaluating, by the one or more processors, the received sensor data in relation to the grid;
determining, by the one or more processors based on evaluating the received sensor data in relation to the grid, whether any cell in the grid contains an obstacle that overlaps with the trajectory; and
the one or more processors causing the vehicle to take a responsive action upon determining that any cell in the grid contains the obstacle that overlaps with the trajectory.
2. The method of claim 1 , wherein a center of the grid is arranged at a selected position relative to a current pose of the vehicle.
3. The method of claim 2 , wherein the center of the grid is maintained at the selected position as the vehicle moves and the current pose of the vehicle changes.
4. The method of claim 1 , wherein each cell of the grid has a fixed size.
5. The method of claim 1 , wherein determining whether any cell in the grid contains an obstacle that overlaps with the trajectory includes assigning a probability to each cell that that cell is occupied by an obstacle, based on the received sensor data.
6. The method of claim 5 , further comprising updating, as the vehicle moves along the trajectory, the probability assigned to each cell based on updated sensor data received from the perception system.
7. The method of claim 5 , wherein the probability is associated with an occupancy value.
8. The method of claim 7 , further comprising determining that a given cell contains a static object when the occupancy value of the given cell exceeds a threshold.
9. The method of claim 1 , wherein new cells are added to the grid as the vehicle moves along the trajectory.
10. The method of claim 9 , wherein each new cell is initialized with a value indicating that it is unknown whether that cell is occupied by an obstacle.
11. The method of claim 1 , wherein the grid is a three-dimensional grid.
12. The method of claim 1 , wherein each cell of the grid has a same shape.
13. The method of claim 1 , further comprising verifying the trajectory upon determining that no cell in the grid contains an obstacle that overlaps with the trajectory.
14. A method comprising:
determining, by one or more processors, a projected trajectory of a vehicle operating in an autonomous driving mode, the trajectory identifying proposed movement of the vehicle;
performing, by the one or more processors, a simulation of the vehicle moving along the projected trajectory and coming to a stop, the simulation following the projected trajectory for a duration of at least one update cycle of a static occupancy grid, the static occupancy grid having a set of cells in which each cell is configured to store an indication of whether an obstacle in present;
determining, by the one or more processors, whether any cell in the static occupancy grid contains an indication of a presence of an obstacle that would overlap with the projected trajectory; and
upon determining that no cell in the static occupancy grid contains the indication of the presence of an obstacle, the one or more processors causing the vehicle to follow the projected trajectory.
15. The method of claim 14 , wherein the projected trajectory has a width which is at least as large as a width of the vehicle.
16. The method of claim 14 , wherein the update cycle is at least 100 milliseconds.
17. The method of claim 14 , wherein simulating coming to a stop includes simulating application of at least one of maximum jerk and maximum deceleration.
18. The method of claim 14 , wherein determining whether any cell in the static occupancy grid contains an indication of the presence of an obstacle that would overlap with the projected trajectory includes sampling a projected pose of the vehicle at either predetermined times or predetermined distances along the projected trajectory.
19. The method of claim 14 , wherein upon determining that at least one cell in the static occupancy grid contains the indication of the presence of an obstacle, the one or more processors either (i) publishing an alert message or (ii) sending a signal to a speed control module of the vehicle in order to cause the vehicle to slow down or stop.
20. The method of claim 14 , wherein determining whether any cell in the static occupancy grid contains an indication of the presence of an obstacle that would overlap with the projected trajectory includes evaluating whether a detected obstacle would overlap with a buffered pose of the vehicle.