Data Sheet

Page 7
The principle of the obstacle selection is as follows: CE30-A selects the nearest obstacle to LiDAR. Then it gathers
detecting and computing resources, to calculate the azimuth and the projected distance of the obstacle more precisely.
Figure 8 Testing Data Illustration (Top View)
Among them, azimuth represents the angular deviation between the obstacle and the central line of the LiDAR’s HFOV
(0 for central line, negative for left and positive for right) and the degree of trajectory deflection required to avoid the
obstacle.
Projected distance represents the projected distance from the obstacle to the robot (the robot is square at default and
LiDAR is installed on the front surface of the robot). It also indicates the emergence level of avoiding the obstacle.
3.2. Setting Warning Region in Obstacle Avoidance Mode
In many cases, not all objectives within the FOV worth the warning or the response of obstacle avoidance. The warning
region can be set up in the obstacle avoidance mode. Once it is set up, though the detection range is the same, only the
information of the obstacles in the warning region will be reported to the robot.
We could set the region of interest (ROI) by the width and depth:
1) Width: the width of a LiDAR-centered area extending symmetrically. It’s usually the same as that of the robot, i.e.
the width of the robot in the direction of forward motion.
2) Depth: the projected distance to the LiDAR, for which the front surface of LiDAR is set to be the zero plane.
Usually, it corresponds to the distance that the robot needs to make brake in response to obstacles.
Figure 9 Schematic Diagram of ROI Function Description
In the obstacle avoidance mode with ROI setting, CE30-A will preferentially trace the obstacles in the ROI. For example,
Objective A inside the ROI and Objective B outside the ROI exist simultaneously. Even though Objective B is nearer to
CE30-A than Objective A, CE30-A still returns the information of Objective A rather than Objective B, as shown in
Figure 10.