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 input data comprising sensor data from a sensor suite onboard the vehicle, an ego-vehicle state of the vehicle, and an environmental representation which identifies a set of dynamic agents;
based on the input data:
determining a primary plan at an autonomous computing system; and
determining a secondary plan, comprising:
determining a set of navigational edge candidates;
classifying a navigational edge candidate as available based on an occupancy prediction corresponding to each dynamic agent, wherein each dynamic agent is assumed to be static within a planning horizon; and
based on the classification of the navigational edge as available, generating the secondary plan based on the ego-vehicle state and the navigational edge candidate;
storing the secondary plan at a memory coupled to an embedded controller;
autonomously controlling the vehicle based on the primary plan; and
while autonomously controlling the vehicle based on the primary plan, determining satisfaction of a trigger condition and, in response, autonomously controlling the vehicle based on the secondary plan with the embedded controller.
2. The method of claim 1 , wherein the method further comprises: while controlling the vehicle based on the primary plan, updating the secondary plan based on a satisfaction of an expiration condition of the secondary plan.
3. The method of claim 1 , further comprising, while controlling the vehicle based on the secondary plan:
estimating an ego-vehicle pose using odometry;
based on the ego-vehicle pose with a set of secondary sensors, detecting an obstacle along a path of the vehicle corresponding to the secondary plan; and
executing a full stop based on the obstacle detection.
4. The method of claim 3 , wherein the set of secondary sensors comprises time-of-flight sensors of the vehicle sensor suite.
5. The method of claim 1 , wherein the available state classification is based on a map.
6. The method of claim 1 , wherein the available state classification is based on an ego-vehicle dimension and a satisfaction of a minimum passing clearance.
7. The 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.
8. The method of claim 1 , further comprising: selecting the navigational edge from a plurality of available navigational edges based on at least one of: collision constraints, kinematic constraints, and a minimum distance constraint.
9. The method of claim 1 , wherein the trigger condition comprises a communication lapse between the autonomous computing system and a validation endpoint.
10. The method of claim 9 , wherein the validation endpoint comprises a vehicle communication network node.
11. The method of claim 1 , 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.
12. The method of claim 11 , wherein autonomously controlling the vehicle based on the primary plan comprises providing a trajectory to a vehicle control system via the low-level safety platform and the vehicle communication network.
13. A system for a vehicle comprising:
a first computing system communicatively coupled to a first set of vehicle sensors, the first computing system configured to:
based on measurements from the first set of vehicle sensors, determine an ego-vehicle state estimate and an environmental representation which identifies a set of dynamic agents;
determine a primary plan based on the ego-vehicle state estimate and the environmental representation; and
determine a secondary plan, comprising:
determining a set of navigational edges;
classifying a navigational edge as an available state based on an occupancy prediction corresponding to each dynamic agent, wherein each dynamic agent is assumed to be static within a planning horizon; and
based on the classification of the navigational edge as an available state, generating the secondary plan based on the ego-vehicle state and the navigational edge; and
a second computing system comprising an embedded controller and configured to:
receive the secondary plans from the first computing system;
store the secondary plan in a memory of the embedded controller; and
in response to determining satisfaction of a trigger condition, autonomously control the vehicle based on the secondary plan using the embedded controller and a second set of sensors.
14. The system of claim 13 , wherein the second computing system autonomously controls the vehicle based on the primary plan by selectively relaying the primary plan to a vehicle control system via a vehicle communication network.
15. The system of claim 13 , wherein the first computing system is further configured to update the secondary plan based on a satisfaction of an expiration condition of the secondary plan while the vehicle is controlled based on the primary plan.
16. The system of claim 13 , wherein the secondary plan is associated with a trajectory, wherein autonomously controlling the vehicle based on the secondary plan comprises:
estimating an ego-vehicle pose using odometry;
detecting an obstacle along the trajectory based on the ego-vehicle pose with the second set of sensors; and
facilitate execution of a hard stop based on the obstacle detection.
17. The system of claim 13 , wherein the available state classification is based on an ego-vehicle dimension and a satisfaction of a minimum passing clearance.