Robot anti-falling detection method

Abstract

The present invention provides a robot anti-falling detection method. The Kalman filtering algorithm is adopted for optimal estimation, an estimation structure is updated through multiple times of sampling and iterative calculation, and therefore errors caused by data uncertainty of an ultrasonic sensor are eliminated, work is stable, and result reliability is high. The whole control method is simple in structure and concise in algorithm, is suitable for the characteristics of a parallel processing structure and bottom layer design of the FPGA, is higher in operation speed, and can achieve thereal-time monitoring. By adopting a rotary scanning mode of the ultrasonic sensor, the cost is lower, the implementation is simple, steps can be detected in advance and early warned, and the robot iseffectively prevented from falling off.

Description

technical field

[0001] The invention relates to the technical field of robot drop detection, in particular to a robot drop prevention detection method. Background technique

[0002] The main method used by traditional robots to prevent falling is to install ultrasonic, infrared, laser and other distance measuring sensors vertically downward on the bottom of the front of the robot, and measure the falling distance downward. When the table or steps are protruded, the distance changes. For the imminent fall, stop going. This method needs to place the sensor at the position protruding from the tabletop and steps. Only when it has reached the edge can it detect and stop advancing. Because the sensor is unstable and the detection error prevention margin is small, there is a risk of falling due to misjudgment . Contents of the invention

[0003] In order to overcome the shortcomings of the above technologies, the present invention provides a detection method that can detect ste...

Images