Multi-robot co-location algorithm based on square root cubature Kalman filtering (SR-CKF)

Abstract

The invention discloses a multi-robot co-location algorithm based on square root cubature Kalman filtering (SR-CKF), and belongs to the field of the co-location of robots. The whole algorithm is divided into two steps: prediction and updating. First, motion equations and observation equations of robots are established, a relative azimuth is utilized as a measured value, and a multi-robot co-location dynamic model is further obtained. A prediction stage comprises the following steps of calculating a volume point set; propagating volume points through a state equation; pre-estimating pose states of the robots and predicting a square root factor. An updating stage comprises the following steps of calculating the volume point set; calculating a Kalman gain; calculating pose information and the square root factor; updating a state vector, a covariance matrix and the pose information. According to the multi-robot co-location algorithm based on the SR-CKF, which is provided by the embodiment of the invention, a target state mean value and the square root factor of the covariance matrix are directly transmitted in an updating process; the calculation complexity is decreased; the symmetrical characteristic and the positive semi-definite characteristic of the covariance matrix are ensured; the precision and the stability of a numerical value are improved.

Description

technical field [0001] The invention relates to the field of robot cooperative positioning, and relates to a multi-robot cooperative positioning algorithm based on square root volumetric Kalman filtering. Background technique [0002] As people's research fields continue to expand, extending from land and ocean to deep sea and even unknown outer space, multi-robot collaborative systems are playing an increasingly important role in military, aerospace, service, industry and other fields. Being able to perceive, model and determine its own position on the unknown and complex external environment information is the premise and basis for the robot's autonomous navigation. Multi-robot co-localization refers to mutual observation between multiple robots, independent of the external environment, by sharing environmental information, and realizing the determination of their respective pose information in a common environment. [0003] At present, there are many methods to solve mul...

AI Technical Summary

Problems solved by technology

This algorithm has better real-time performance, but the algorithm is prone to error accumulation, and its calculation amount will increase sharply with the increase of the map

Images