Fast Image Processing and Flexible Path Generation System for RoboCup Small Size League

Abstract

Two key features of successful multi-robot systems in RoboCup are robustness of a vision system and optimization of feedback control in order to reach the goal point and generate the action under the team strategy. This paper proposes two new methods. One is a fast image processing method, which is coped with the spatial variance of color parameters in the field, to extract the positions of robots and ball in 1/30 sec. The separation problem in the interlaced format image is solved. Another one is a path generation method in which the robot approaches the goal by changing its direction convergently. With these two algorithms, the real time processing system is realized by generating a stable path under a low quality input image.

Robust and Accurate Detection of Object Orientation and ID Without Color Segmentation

“…Geometric asymmetry can be used to find the rotational correspondence for orientation estimation. Pattern (c), called "pie slice-based", is unique and is described in (S. Hibino et al, 2002). The method that uses this patch scans the circular pattern using markers on the robot.…”

Cooperative Soccer Play by Real Small-Size Robot

“…Figure 2 shows the flowchart of image processing. The outline of processing is shown below and literatures [6][7][8][9] give detailed description.…”

“…Step-4 (Labeling) Apply the first propagation in the labeling algorithm, which we have developed last year [9], to label the object region. Figure 5 shows the result.…”

Fast and accurate vision-based pattern detection and identification

“…One alternative to this is to have a key patch marking the center of the pattern, surrounded by radial "pie slices" of two or more colors. Orientation and identification can be performed by scanning at some constant radius from the central patch [7]. Unfortunately, by depending on features (color edges in this case) that are difficult to quickly detect with sub-pixel accuracy, it is difficult to get very accurate orientation using this method.…”

“…Thus we feel this is a good testbed for a tracking system. Two other vision tracking and identification systems for the F180 league have been described in [9] and [7]. Each describes a working system used by a team, but neither motivates the choice of pattern by a thorough analysis of the underlying feature error, or attempts to generalize detection to other similar patterns in order to compare their performance.…”