Robotic Intelligence and Automation最新文献

Purpose

This paper aims to present a novel transfer alignment method based on combined double-time observations with velocity and attitude for ships’ poor maneuverability to address the system errors introduced by flexural deformation and installing which are difficult to calibrate.

Design/methodology/approach

Based on velocity and attitude matching, redesigning and deducing Kalman filter model by combining double-time observation. By introducing the sampling of the previous update cycle of the strapdown inertial navigation system (SINS), current observation subtracts previous observation are used as measurements for transfer alignment filter, system error in measurement introduced by deformation and installing can be effectively removed.

Findings

The results of simulations and turntable tests show that when there is a system error, the proposed method can improve alignment accuracy, shorten the alignment process and not require any active maneuvers or additional sensor equipment.

Originality/value

Calibrating those deformations and installing errors during transfer alignment need special maneuvers along different axes, which is difficult to fulfill for ships’ poor maneuverability. Without additional sensor equipment and active maneuvers, the system errors in attitude measurement can be eliminated by the proposed algorithms, meanwhile improving the accuracy of the shipboard SINS transfer alignment.

Purpose

This paper mainly aims to deal with the problems of uncertainties including modelling errors, unknown dynamics and disturbances caused by load mutation in control of permanent magnet synchronous motor (PMSM).

Design/methodology/approach

This paper proposes an enhanced speed sensorless vector control method based on an active disturbance rejection controller (ADRC) for a PMSM. First, a state space model of the PMSM is obtained for the field orientation control strategy. Second, a sliding mode observer (SMO) based on back electromotive force (EMF) is introduced to replace the encode to estimate the rotor flux position angle and speed. Third, an infinite impulse response (IIR) filter is introduced to eliminate high frequency noise mixed in the output of the sliding mode observer. In addition, a speed control method based on an extended state observer (ESO) is proposed to estimate and compensate for the total disturbances. Finally, an experimental set-up is built to verify the effectiveness and superiority of the proposed ADRC-based control method.

Findings

The comparative experimental results show that the proposed speed sensorless control method with the IIR filter can achieve excellent robustness and speed tracking performance for PMSM system.

Research limitations/implications

An enhanced sensorless control method based on active disturbance rejection controller is designed to realize high precision control of the PMSM; the IIR filter is used to attenuate the chattering problem of traditional SMO; this method simplifies the system and saves the total cost due to the speed sensorless technology.

Practical implications

The use of sensorless can reduce costs and be more beneficial to actual industrial application.

Originality/value

The proposed enhanced speed sensorless vector control method based on an ADRC with the IIR filter enriches the control method of PMSM. It can ameliorate system robustness and achieve excellent speed tracking performance.

Purpose

This paper aims to focus on lane detection of unmanned mobile robots. For the mobile robot, it is undesirable to spend lots of time detecting the lane. So quickly detecting the lane in a complex environment such as poor illumination and shadows becomes a challenge.

Design/methodology/approach

A new learning framework based on an integration of extreme learning machine (ELM) and an inception structure named multiscale ELM is proposed, making full use of the advantages that ELM has faster convergence and convolutional neural network could extract local features in different scales. The proposed architecture is divided into two main components: self-taught feature extraction by ELM with the convolution layer and bottom-up information classification based on the feature constraint. To overcome the disadvantages of poor performance under complex conditions such as shadows and illumination, this paper mainly solves four problems: local features learning: replaced the fully connected layer, the convolutional layer is used to extract local features; feature extraction in different scales: the integration of ELM and inception structure improves the parameters learning speed, but it also achieves spatial interactivity in different scales; and the validity of the training database: a method how to find a training data set is proposed.

Findings

Experimental results on various data sets reveal that the proposed algorithm effectively improves performance under complex conditions. In the actual environment, experimental results tested by the robot platform named BIT-NAZA show that the proposed algorithm achieves better performance and reliability.

Originality/value

This research can provide a theoretical and engineering basis for lane detection on unmanned robots.

Purpose

The purpose of this paper is to propose a new acceleration/deceleration (acc/dec) algorithm for motion profiles. The motion efficiency, flexibility of the motion profiles and the residual vibration of the movement are discussed in this paper.

Design/methodology/approach

A dynamics model is developed to assess the residual vibration of these two kinds of motion profile. And a Simulink model is created to assess the motion efficiency and flexibility of the motion profiles with the proposed acc/dec algorithm.

Findings

Considering the flexibility of trigonometric motion profiles and the higher motion efficiency of S-curve motion profiles, the authors add the polynomial parts into the jerk profile of the cosine function acc/dec algorithm to hold the jerk when it reaches the maximum so that the motion efficiency can increase and decrease residual vibration at the same time. And the cyclical parameter k shows the decisive factor for the flexibility of trigonometric motion profiles.

Originality/value

Comparing with the traditional motion profiles, the proposed motion profiles have higher motion efficiency and excite less residual vibration. The acc/dec algorithm proposed in this paper is useful for the present motion control and servo system.

Purpose

Press-fit with force and displacement monitoring is commonly adopted in automotive mechatronic system assembling. However, suitable methods for the press-fit study are still at initial investigation phase. The sequential data physical meaning, small data sets from different resources and computing efficiency should be considered. Therefore, this paper aims to better identify press-fit fault types.

Design/methodology/approach

This paper proposed one-dimensional convolutional neural network (1DCNN)–long short-term memory (LSTM) method to perform press-fit fault diagnosis into automotive assembling practice which is in accordance with current product development procedure. Specialized data augmentation method is proposed to merge different data resources and increase the sample size. Referring one-way sequential data characteristics, LSTM and batch normalization layers are integrated in 1DCNN to improve the performance.

Findings

The proposed 1DCNN-LSTM method is feasible with small data sets from different sources. Using data augmentation to make data unified and sample size increased, the accuracy could reach more than 99%. Training time has reduced from 90 s/Epoch to 4 s/Epoch compare to pure LSTM method.

Originality/value

The proposed method shows better performance with less training time compared to LSTM. Therefore, the method has practical value and is worthy of industrial application.

Purpose

The purpose of this paper is to provide a fast and accurate network for spatiotemporal action localization in videos. It detects human actions both in time and space simultaneously in real-time, which is applicable in real-world scenarios such as safety monitoring and collaborative assembly.

Design/methodology/approach

This paper design an end-to-end deep learning network called collaborator only watch once (COWO). COWO recognizes the ongoing human activities in real-time with enhanced accuracy. COWO inherits from the architecture of you only watch once (YOWO), known to be the best performing network for online action localization to date, but with three major structural modifications: COWO enhances the intraclass compactness and enlarges the interclass separability in the feature level. A new correlation channel fusion and attention mechanism are designed based on the Pearson correlation coefficient. Accordingly, a correction loss function is designed. This function minimizes the same class distance and enhances the intraclass compactness. Use a probabilistic K-means clustering technique for selecting the initial seed points. The idea behind this is that the initial distance between cluster centers should be as considerable as possible. CIOU regression loss function is applied instead of the Smooth L1 loss function to help the model converge stably.

Findings

COWO outperforms the original YOWO with improvements of frame mAP 3% and 2.1% at a speed of 35.12 fps. Compared with the two-stream, T-CNN, C3D, the improvement is about 5% and 14.5% when applied to J-HMDB-21, UCF101-24 and AGOT data sets.

Originality/value

COWO extends more flexibility for assembly scenarios as it perceives spatiotemporal human actions in real-time. It contributes to many real-world scenarios such as safety monitoring and collaborative assembly.