A method of tracking and positioning of industrial robots

: In this paper, an improved tracking and localization algorithm of an omni-directional mobile industrial robot is proposed to meet the high positional accuracy requirement, improve the robot’s repeatability positioning precision in the traditional trilateral algorithm, and solve the problem of pose lost in the moving process. Laser sensors are used to identify the reflectors, and by associating the reflectors identified at a particular time with the reflectors at a previous time, an optimal triangular positioning method is applied to realize the positioning and tracking of the robot. The experimental results show that positioning accuracy can be satisfied, and the repeatability and anti-jamming ability of the omni-directional mobile industrial robot will be greatly improved via this algorithm.


Introduction
Across many domains, there is an increasing demand for robots capable of performing complex and dexterous manipulation tasks. A typical example is the need for factory assembly lines. With the combination of a dexter-ous manipulator and mobile platforms, mobile robots are well suited to these complex tasks. Given any complex scenario, the mobile robot must be able to achieve localization (Suwoyo, Abdurohman, et al., 2022;Yu et al., 2023). Localization is a key functionality of any navigation system as it tracks and determines the position of a mobile robot in the environment. This is a challenging topic in the area of autonomous mobile robot research (Akbar Qureshi et al., 2022;J. Zhang et al., 2023).
Many methods are proposed to address the problem of mobile robot localization; these methods can be divided as two categories: relative positioning and absolute positioning. In relative positioning, dead reckoning and inertial navigation are commonly used to calculate mobile robots' positions. This method does not need to perceive the external environment, but its drift error accumulates over time. Absolute positioning relies on detecting and recognizing different features in the environment in order to realize the position of the robot. These environment features are normally divided into two types: artificial landmarks and natural landmarks (Suwoyo & Harris Kristanto, 2022). Compared with the natural landmarks, artificial landmarks have advantages of high recognition and high accuracy. There is no cumulative error when a localization method based on artificial landmarks is utilized; however, the key challenge is accurately identifying and extracting the needed information from the artificial landmarks. Researchers have proposed several solutions, such as fuzzy reflector-based localization and color reflector-based self-localization. The main drawbacks of these methods are that the amount of computation power they need is too large and the antiinterference ability is poor (Suwoyo, Hidayat, et al., 2022;Tai & Yeung, 2022).
To tackle these problems pertaining to the localization method based on artificial landmarks, Madsen and Andersen proposed a three reflectors positioning method using the constraints of three reflectors and the triangulation principle to realize positioning. Betke and Gurvits proposed a multi-channel positioning method with the three-dimensional positioning principle and the least squares method to accomplish localization. Because of the unavoidable errors in position and angle measurement of reflectors, use of only a trilateral or triangular method will not achieve the required positioning accuracy (Fan et al., 2023). In the navigation of a mobile robot, some reflectors may be obscured or confused with other obstacles, leading to pose information loss of the positioning system. To solve these problems, this paper presents an improved tracking and locating algorithm of an omni-directional mobile industrial robot. The robot is used in drilling and riveting processing of sheet metal parts, and laser sensors are used to identify the reflectors. By associating the reflectors identified at a particular moment with the reflectors at a previous time, an optimal triangular positioning method is applied to realize the positioning and tracking of the robot in a global environment (Suryaprakash et al., 2021;Tan et al., 2023).

Research Method
Optimal triangulation positioning algorithm based on angle measurement In order to achieve accurate positioning, a mobile robot must have a basic ability to perceive external information through extracting the reflector features (Zhang et al., 2022). In this research, there are n reflectors, and the feature of each reflector Bi (i = 1,2,•••, n) is extracted from the measurement data. The feature extraction algorithm consists of three steps: (i) filtering and clustering, (ii) identification, and (iii) feature extraction.

Coordinate system description
The measurement data obtained by each scan cycle of the laser sensor are a set of discrete data sequences {{( , ), }i | = 1,2, ⋯ , }, where ( , )i is the polar coordinate, the distance from the target point to the sensor, the polar angle, and λi the intensity value of the ith data point.
In the process of data analysis, the outlier points that are contaminated by noise need to be filtered out (D. Zhang et al., 2023).
Considering that the density of the collected data points is proportional to the distance from the target point to the laser sensor, and to improve the efficiency of the feature extraction process, an adaptive clustering method is adopted. Unless the distance between two data points is less than the threshold d, these data points are clustered for one reflector (Li et al., 2023;Ostanin et al., 2022). (1) where i-1 is the distance value of the − 1th data point, ∆ the angle resolution of the laser sensor, β an auxiliary constant parameter, and y the measurement error. The values of parameters y and β are given as 0.01 m and 10°, respectively.

Result and Discussion
The experimental data are obtained by the LMS 100 laser sensor with a scanning range of 270° and an angular resolution of 0.25°. The outside of the reflector is wrapped by reflective tape.

Repeatability positioning results
The optimal triangulation method based on the angle measurement is used for validation by the repeatability of the omni-directional mobile industrial robot. In the stationary state of the omni-directional mobile industrial robot, the environment is scanned by laser sensors to realize the positioning of the robot. Each positioning result is indicated by a red dot in Fig. 3.
The repeatability obtained by the trilateral method is nearly 18 mm, while the repeatability of the optimal method is only 9 mm. It can be shown that the optimal method is better than the traditional method (Sun et al., 2023).

Positioning accuracy results
The omni-directional mobile industrial robot moves in the direction of the arrow in Fig. 3, and each time the robot moves a certain distance, the navigation and positioning system will perform a positioning experiment, i.e., it will use the left rear laser sensor to calculate the current position. An average of 30 samples is taken for each experiment. Figure 4 shows the results of static positioning accuracy. The maximum distance error is 18 mm and the maximum angle error is 2°, which satisfies the positioning requirement of the omnidirectional mobile industrial robot.

Tracking location results
The omni-directional mobile industrial robot moves in the designated route, and it needs to constantly calculate and record its own positioning in the moving process. As shown in Fig. 5, the trajectory of the moving robot based on the tracking and positioning method is smoother.

Conclusions
This paper demonstrates the feasibility of a tracking and locating algorithm for omnidirectional mobile industrial robot. The following conclusions can be drawn from this study: (i) In the detection of a reflector in the sensor coordinate system, the angle repeatability of the reflector is better than that of the distance repeatability based on the feature extraction algorithm; (ii) The repeatability positioning accuracy using the optimal triangulation method based on the angle measurement is nearly 9 mm, which is better than that of the trilateral method; (iii) The positioning error of the robot is 18 mm, which satisfies the positioning requirement of omni-directional mobile industrial robot. Improvements in the location method based on reflectors, such as optimizing the layout of reflectors and the map of reflectors selection strategy for positioning, are still needed. In the future, we intend to extend our work to research a positioning method based on reflectors and the environmental profile, and achieve better accuracy on the localization of the mobile robot.