Moving robot with improved identification accuracy of step distance
There is provided a moving robot including a light projector, an image sensor and a processing unit. The light projector projects a vertical light segment toward a moving direction. The image sensor captures, toward the moving direction, an image frame containing a light segment image associated with the vertical light segment. The processing unit calculates a step distance and a segment feature according to the image frame, outputs a flag signal according to the segment feature to indicate whether the calculated step distance is confident or not, and perform a pixel interpolation in calculating the step distance to improve the identification accuracy.
1. A moving robot, comprising:
a light projector configured to project a vertical light segment toward a moving direction;
an image sensor configured to capture an image frame containing a light segment image associated with the vertical light segment; and
a first processing unit configured to
calculate an image feature according to the light segment image in the image frame,
output a flag signal according to the image feature to indicate a confidence level of the image frame,
identify a segment height of the light segment image,
calculate and output a current step distance according to the segment height when the flag signal indicates a high confidence level, and
not output the current step distance when the flag signal indicates a low confidence level.
2. The moving robot as claimed in claim 1 , wherein the first processing unit is arranged inside a sensing chip which comprises an independent leg configured to output the flag signal.
3. The moving robot as claimed in claim 1 , wherein the image feature comprises a segment width of the light segment image and whether the light segment image is a broken line.
4. The moving robot as claimed in claim 1 , further comprising:
a memory configured to previously store a look up table regarding a relationship between multiple segment heights of the light segment image and multiple step distances; and
a second processing unit configured to access the look up table to identify another current step distance when the flag signal indicates the low confidence level.
5. The moving robot as claimed in claim 1 , further comprising:
a second sensor, which is not an image sensor; and
a second processing unit configured to activate the second sensor to detect another current step distance when the flag signal indicates the low confidence level.
6. The moving robot as claimed in claim 1 , wherein the first processing unit is further configured to
not calculate the current step distance when the flag signal indicates the low confidence level.
7. The moving robot as claimed in claim 1 , wherein the first processing unit is further configured to
calculate gray level sums of every pixel column of multiple pixel columns of the image frame,
identify two pixel columns having the gray level sums closest to a predetermined gray level threshold, and
perform a pixel interpolation on the two pixel columns and two pixel rows closest to the predetermined gray level threshold.
8. The moving robot as claimed in claim 1 , further comprising another light projector configured to project another vertical light segment toward the moving direction, and the first processing unit is further configured to
digitize the image frame by a digitizing threshold,
divide the image frame into a left image frame and a right image frame respectively containing one light segment image, and
respectively calculate an image feature of the one light segment image in the left image frame and the right image frame.
9. A moving robot, comprising:
a light projector configured to project a vertical light segment toward a moving direction;
an image sensor configured to capture an image frame containing a light segment image associated with the vertical light segment;
another light projector configured to project another vertical light segment toward the moving direction; and
a first processing unit configured to
digitize the image frame by a digitizing threshold,
divide the image frame into a left image frame and a right image frame respectively containing one light segment image,
respectively calculate an image feature and a step distance according to the one light segment image in the left image frame and the right image frame, and
output the step distance and a flag signal indicating a confidence level of the step distance.
10. The moving robot as claimed in claim 9 , wherein the first processing unit is arranged inside a sensing chip which comprises a leg configured to output the flag signal but not output the step distance.
11. The moving robot as claimed in claim 9 , wherein the image feature comprises a segment width of the one light segment image and whether the one light segment image is a broken line.
12. The moving robot as claimed in claim 9 , further comprising:
a memory configured to previously store a look up table regarding a relationship between multiple segment heights of the one light segment image and multiple step distances; and
a second processing unit configured to access the look up table to identify a current step distance and ignore the step distance outputted by the first processing unit when the flag signal indicates a low confidence level.
13. The moving robot as claimed in claim 9 , further comprising:
a second sensor, which is not an image sensor; and
a second processing unit configured to activate the second sensor to detect a current step distance and ignore the step distance outputted by the first processing unit when the flag signal indicates a low confidence level.
14. The moving robot as claimed in claim 9 , wherein the first processing unit is further configured to
calculate gray level sums of every pixel column of multiple pixel columns of the image frame,
identify two pixel columns having the gray level sums closest to a predetermined gray level threshold, and
perform a pixel interpolation on the two pixel columns and two pixel rows closest to the predetermined gray level threshold.
15. A moving robot, comprising:
a memory configured to store a gray level threshold;
a light projector configured to project a vertical light segment toward a moving direction;
an image sensor configured to capture an image frame containing a light segment image associated with the vertical light segment; and
a processing unit configured to
calculate gray level sums of every pixel column of multiple pixel columns of the image frame,
identify two pixel columns having the gray level sums closest to the gray level threshold,
perform a pixel interpolation on the two pixel columns and two pixel rows closest to the gray level threshold, and
calculate a current step distance according to a segment height corresponding to an interpolated subpixel column position.
16. The moving robot as claimed in claim 15 , wherein the gray level is
a predetermined fixed value, or
a varied value determined according to the gray level sum of one pixel column not containing the light segment image.
17. The moving robot as claimed in claim 15 , wherein the two pixel columns correspond to a front end of the light segment image.
18. The moving robot as claimed in claim 15 , wherein the segment height is a number of pixel rows containing the light segment image.
19. A moving robot, comprising:
a light projector configured to project a light stripe toward a moving direction;
an image sensor configured to capture an image frame containing a light stripe image associated with the light stripe;
a second sensor; and
a processing unit configured to calculate an image feature according to the light stripe image in the image frame to accordingly output a flag signal,
wherein the moving robot is configured to use the second sensor to sense an area toward the moving direction when the flag signal indicates a low confidence level.