Autonomous vehicle safety platform system and method
A system 100 for autonomous vehicle operation can include: a low-level safety platform 130 ; and can optionally include and/or interface with any or all of: an autonomous agent 102 , a sensor system, a computing system 120 , a vehicle communication network 140 , a vehicle control system 150 , and/or any suitable components. The system functions to facilitate fallback planning and/or execution at the autonomous agent. Additionally or alternatively, the system can function to transition the autonomous agent between a primary (autonomous) operation mode and a fallback operation mode.
1. An autonomous fallback method for a vehicle, comprising:
receiving an environmental representation which identifies a set of dynamic agents;
based on the environmental representation, determining a set of navigational edge candidates;
classifying a navigational edge candidate of the set of navigational edge candidates as available based on an occupancy prediction corresponding to each dynamic agent of the set of dynamic agents, wherein each dynamic agent is assumed to be static within a planning horizon;
based on the classification of the navigational edge candidate as available, generating a fallback plan based on the navigational edge candidate;
storing the fallback plan at a memory coupled to an embedded controller; and
determining satisfaction of a trigger condition and, in response, autonomously controlling the vehicle based on the fallback plan with the embedded controller.
2. The autonomous fallback method of claim 1 , further comprising: updating the fallback plan based on a satisfaction of an expiration condition associated with the fallback plan.
3. The autonomous fallback method of claim 2 , wherein the expiration condition comprises a detection that a second fallback plan associated with a second point in time has been generated, wherein the second point in time is later than a first point in time associated with the fallback plan.
4. The autonomous fallback method of claim 1 , while autonomously controlling the vehicle based on the fallback plan:
with vehicle odometry, traversing a path towards the navigational edge candidate;
detecting an obstacle along the path of the vehicle; and
executing a full stop based on the obstacle detection.
5. The autonomous fallback method of claim 1 , wherein the available classification of the navigational edge candidate is based on a map.
6. The autonomous fallback method of claim 1 , wherein the set of navigational edge candidates are determined at a predetermined set of distances relative to an ego-vehicle position.
7. The autonomous fallback method of claim 1 , further comprising: selecting the navigational edge from a plurality of available navigational edges based on a kinematic constraint.
8. The autonomous fallback method of claim 1 , further comprising: selecting the navigational edge from a plurality of available navigational edges based on a minimum distance constraint.
9. The autonomous fallback method of claim 1 , wherein the fallback plan is determined with an autonomous computing system of the vehicle, wherein the trigger condition comprises a communication failure of the autonomous computing system.
10. The autonomous fallback method of claim 9 , wherein the embedded controller is within a low-level safety platform which is communicatively connected to the autonomous computing system and a vehicle communication network.
11. The autonomous fallback method of claim 1 , further comprising: prior to determining satisfaction of the trigger condition, autonomously controlling the vehicle based on a primary plan, wherein the primary plan is determined based on the environmental representation.
12. The autonomous fallback method of claim 11 , wherein the primary plan and fallback plan are each determined with a high-level planner, wherein the embedded controller is within a low-level safety platform, wherein the trigger condition comprises a communication failure between the high-level planner and the low-level safety platform.
13. The autonomous fallback method of claim 1 , wherein the set of dynamic agents comprises an automobile.
14. A method for a vehicle, comprising:
determining an environmental representation comprising a set of dynamic agents;
determining a set of navigation candidates based on the environmental representation;
classifying a navigation candidate based on an occupancy prediction corresponding to each dynamic agent, wherein the occupancy prediction for each dynamic agent is fixed within a planning horizon;
based on the classification of the navigation candidate, generating a fallback plan based on the navigation candidate;
determining satisfaction of a trigger condition; and
in response to determining satisfaction of the trigger condition, autonomously controlling the vehicle with an embedded controller based on the fallback plan.
15. The method of claim 14 , further comprising: updating the fallback plan based on a satisfaction of an expiration condition associated with the fallback plan.
16. The method of claim 14 , further comprising, while autonomously controlling the vehicle, based on the fallback plan:
traversing a path towards the navigation candidate; and
executing a full stop based on an obstacle detection along the path.
17. The method of claim 14 , wherein the set of navigation candidates are determined at a predetermined set of distances relative to an ego-vehicle position.
18. The method of claim 14 , further comprising: selecting the navigation candidate from a plurality of available navigation candidates based on a kinematic constraint or a minimum distance constraint.
19. The method of claim 14 , wherein the fallback plan is determined with an autonomous computing system of the vehicle, wherein the trigger condition comprises a communication failure of the autonomous computing system.
20. The method of claim 14 , wherein the embedded controller is within a low-level safety platform which is communicatively connected to an autonomous computing system and a vehicle communication network.