Determining speeds for an autonomous vehicle

1. A method, comprising:

determining a first reference line representing a path through an environment for an autonomous driving vehicle;

determining a speed constraint function based on a set of speed limits associated with the environment, a set of curvatures of the path, and a set of obstacles in the environment, wherein the speed constraint function comprises a continuous function, and wherein the determining the speed constraint function comprises discretizing one or more of the set of speed limits associated with the environment, the set of curvatures of the path, and the set of obstacles in the environment, based on a specified resolution, wherein the speed constraint function includes a combination of a set of polynomial functions that represent a set of segments, and wherein a constant second term is used to connect consecutive segments of the set of segments;

determining a set of speeds for the path through the environment based on the speed constraint function; and

controlling the autonomous driving vehicle based on the path and the set of speeds.

2. The method of claim 1 , wherein the determining the speed constraint function further comprises:

determining the set of speeds based on the set of speed limits associated with the environment, the set of curvatures of the path, the set of obstacles in the environment, and the specified resolution.

3. The method of claim 2 , wherein the determining the speed constraint function further comprises:

forming the set of segments between the set of speeds by connecting each speed in the set of speeds with a next speed in the set of speeds.

4. The method of claim 1 , wherein:

the set of speed limits associated with the environment, the set of curvatures of the path, and the set of obstacles in the environment are represented using a set of functions; and

at least one of the set of functions comprises a discontinuous function.

5. The method of claim 1 , wherein:

the set of speed limits associated with the environment, the set of curvatures of the path, and the set of obstacles in the environment are represented using a set of functions; and

at least one of the set of functions is not differentiable.

6. The method of claim 1 , wherein the speed constraint function comprises a differentiable function.

7. A non-transitory machine-readable medium having instructions stored therein, which when executed by a processor, cause the processor to perform operations, the operations comprising:

determining a first reference line representing a path through an environment for an autonomous driving vehicle;

determining a speed constraint function based on a set of speed limits associated with the environment, a set of curvatures of the path, and a set of obstacles in the environment, wherein the speed constraint function comprises a continuous function, and wherein the determining the speed constraint function comprises discretizing one or more of the set of speed limits associated with the environment, the set of curvatures of the path, and the set of obstacles in the environment, based on a specified resolution, wherein the speed constraint function includes a combination of a set of polynomial functions that represent a set of segments, and wherein a constant second term is used to connect consecutive segments of the set of segments;

determining a set of speeds for the path through the environment based on the speed constraint function; and

controlling the autonomous driving vehicle based on the path and the set of speeds.

8. The non-transitory machine-readable medium of claim 7 , wherein the determining the speed constraint function further comprises:

determining the set of speeds based on the set of speed limits associated with the environment, the set of curvatures of the path, the set of obstacles in the environment, and the specified resolution.

9. The non-transitory machine-readable medium of claim 8 , wherein the determining the speed constraint function further comprises:

forming the set of segments between the set of speeds by connecting each speed in the set of speeds with a next speed in the set of speeds.

10. The non-transitory machine-readable medium of claim 7 , wherein the determining the speed constraint function further comprises:

determining the set of polynomial functions to represent the set of segments, wherein each polynomial function in the set of polynomial functions represents a respective segment in the set of segments.

11. The non-transitory machine-readable medium of claim 10 , wherein the set of polynomial functions comprises a set of quintic polynomial functions.

12. The non-transitory machine-readable medium of claim 10 , wherein the determining the speed constraint function further comprises:

combining the set of polynomial functions to generate the speed constraint function.

13. The non-transitory machine-readable medium of claim 7 , wherein:

the set of speed limits associated with the environment, the set of curvatures of the path, and the set of obstacles in the environment are represented using a set of functions; and

at least one of the set of functions comprises a discontinuous function.

14. The non-transitory machine-readable medium of claim 7 , wherein:

the set of speed limits associated with the environment, the set of curvatures of the path, and the set of obstacles in the environment are represented using a set of functions; and

at least one of the set of functions is not differentiable.

15. The method of claim 9 , wherein the determining the speed constraint function further comprises:

determining the set of polynomial functions to represent the set of segments, wherein each polynomial function in the set of polynomial functions represents a respective segment in the set of segments.

16. The method of claim 15 , wherein the set of polynomial functions comprises a set of quintic polynomial functions.

17. The method of claim 15 , wherein the determining the speed constraint function further comprises:

combining the set of polynomial functions to generate the speed constraint function.

18. A data processing system, comprising:

a processor; and

a memory coupled to the processor to store instructions, which when executed by the processor, cause the processor to perform operations, the operations including:

determining a first reference line representing a path through an environment for an autonomous driving vehicle;

determining a speed constraint function based on a set of speed limits associated with the environment, a set of curvatures of the path, and a set of obstacles in the environment, wherein the speed constraint function comprises a continuous function, and wherein the determining the speed constraint function comprises discretizing one or more of the set of speed limits associated with the environment, the set of curvatures of the path, and the set of obstacles in the environment, based on a specified resolution, wherein the speed constraint function includes a combination of a set of polynomial functions that represent a set of segments, and wherein a constant second term is used to connect consecutive segments of the set of segments;

determining a set of speeds for the path through the environment based on the speed constraint function; and

controlling the autonomous driving vehicle based on the path and the set of speeds.

19. The data processing system of claim 18 , wherein the determining the speed constraint function further comprises:

determining the set of polynomial functions to represent the set of segments, wherein each polynomial function in the set of polynomial functions represents a respective segment in the set of segments.

20. The data processing system of claim 18 , wherein the determining the speed constraint function further comprises:

combining the set of polynomial functions to generate the speed constraint function.