Angle calibration in light detection and ranging system

1. A light detection and ranging (LIDAR) device, comprising:

a light beam emitter operable to emit a light beam;

a prism set positioned in an optical path of the light beam to refract the light beam onto a surface of a surrounding object, the prism set including at least a first prism and a second prism;

a light detector to detect light reflected by the surface of the surrounding object; and

a controller, in communication with the light detector, configured to obtain a plurality of points indicative of a surface shape of the surrounding object based on the detected light,

wherein the controller is operable to (1) evaluate a cost function that accounts for a change of the surface shape of the surrounding object, (2) determine a relative bias in the prism set, the relative bias being a difference between a rotational deviation of the first prism and a rotational deviation of the second prism; and (3) cause, based on the relative bias in the prism set, a compensation for an estimation error in the controller's estimation of the surface of the surrounding object.

2. The device of claim 1 , wherein the determining of the relative bias includes minimizing the cost function.

3. The device of claim 1 , wherein the evaluation of the cost function includes, for each point of the plurality of points, (1) determining a normal vector of the point, the point representing a dot on the surface of the surrounding object as measured by the controller using the detected light; and (2) comparing the normal vector of the point with a normal vector of an adjacent point to the point.

4. The device of claim 3 , wherein the adjacent point is temporally adjacent to the point.

5. The device of claim 3 , wherein the adjacent point is spatially adjacent to the point.

6. The device of claim 3 , wherein the controller is operable to exclude the point in response to a determination that an angle between the normal vector of the point and the normal vector of the adjacent point exceeds a predetermined threshold.

7. The device of claim 1 , wherein the controller is further operable to determine a collective bias in the prism set, the collective bias being a rotational deviation of the prism set.

8. The device of claim 7 , wherein the collective bias is determined based on an evaluation of an estimated normal vector of a reference object.

9. The device of claim 8 , wherein the controller is operable to perform a random sample consensus (RANSAC) function on the plurality of points to identify the reference object.

10. The device of claim 7 , wherein the controller, in causing the compensation for the estimation error, is operable to determine the rotational deviation of the first prism and the rotational deviation of the second prism based on the relative bias and the collective bias.

11. A calibration system for a light detection and ranging (LIDAR) device, comprising:

a memory that stores instructions, and

a processor in communication with the memory and operable to execute the instructions to implement a method of calibrating a prism set of the LIDAR device including at least a first prism and a second prism, the method comprising:

obtaining a plurality of points indicative of a surface shape of a surrounding object;

evaluating a cost function that accounts for a change of the surface shape of the surrounding object using the plurality of points;

determining a relative bias in the prism set using the cost function, wherein the relative bias is a difference between a rotational deviation of the first prism and a rotational deviation of the second prism; and

causing, based on the relative bias, a compensation for an estimation error in an estimation of the surface shape of the surrounding object.

12. The system of claim 11 , wherein the determining of the relative bias includes minimizing the cost function.

13. The system of claim 1 , wherein the evaluating the cost function includes, for each point of the plurality of points:

determining a normal vector of the point, on the surface of the surrounding object as measured by the controller using the detected light; and

comparing the normal vector of the point with a normal vector of an adjacent point to the point.

14. The system of claim 13 , wherein the adjacent point is temporally adjacent to the point.

15. The system of claim 13 , wherein the adjacent point is spatially adjacent to the point.

16. The system of claim 13 , wherein the evaluating of the cost function includes excluding the point in response to a determination that an angle between the normal vector of the point and the normal vector of the adjacent point exceeds a predetermined threshold.

17. The system of claim 11 , further comprising determining a collective bias in the prism set, the collective bias being a rotational deviation of the prism set.

18. The system of claim 17 , wherein the determining the collective bias includes identifying a reference object in the plurality of points.

19. A method of calibrating a light detection and ranging (LIDAR) device, comprising:

obtaining a plurality of points indicative of a surface shape of a surrounding object;

evaluating a cost function that accounts for a change of the surface shape of the surrounding object using the plurality of points;

determining a relative bias between the first prism and the second prism based on the normal vectors of the plurality of points, wherein the relative bias is a difference between a rotational deviation of the first prism and a rotational deviation of the second prism; and

causing, based on the relative bias, a compensation for an estimation error in an estimation of the surface shape of the surrounding object, wherein the determining of the relative bias includes minimizing the cost function.