Search results

Automated robotic assembly on a moving workpiece, referred to as assembly in motion, demands that an assembly robot is synchronised in all degrees of freedom to the moving workpiece, on which assembly parts are installed. Currently, this requirement cannot be met due to the lack of robust estimation of 3D positions and the trajectory of the moving workpiece. The purpose of this paper is to develop a camera system that measures the 3D trajectory of the moving workpiece for robotic assembly in motion.

For the trajectory estimation, an assembly robot-guided, monocular camera system is developed. The motion trajectory of a workpiece is estimated, as the trajectory is considered as a linear combination of trajectory bases, such as discrete cosine transform bases.

The developed camera system for trajectory estimation is tested within the robotic assembly of a cylinder block in motion. The experimental results show that the proposed method is able to reconstruct arbitrary trajectories of an assembly point on a workpiece moving in 3D space.

With the developed technology, a point trajectory can be recovered offline only after all measurement images are acquired. For practical assembly tasks in real production, this method should be extended to determine the trajectory online during the motion of a workpiece.

For practical, robotic assembly in motion, such as assembling tires, wheels and windscreens on conveyed vehicle bodies, the developed technology can be used for positioning a moving workpiece, which is in the distant field of an assembly robot.

Besides laser trackers, indoor global positioning systems and stereo cameras, this paper provides a solution of trajectory estimation by using a monocular camera system.

The purpose of this paper is to present a preliminary solution to address the problem of estimating human pose and trajectory by an aerial robot with a monocular camera in near real time.

The distinguishing feature of the solution is a dynamic classifier selection architecture. Each video frame is corrected for perspective using projective transformation. Then, a silhouette is extracted as a Histogram of Oriented Gradients (HOG). The HOG is then classified using a dynamic classifier. A class is defined as a pose-viewpoint pair, and a total of 64 classes are defined to represent a forward walking and turning gait sequence. The dynamic classifier consists of a Support Vector Machine (SVM) classifier C64 that recognizes all 64 classes, and 64 SVM classifiers that recognize four classes each – these four classes are chosen based on the temporal relationship between them, dictated by the gait sequence.

The solution provides three main advantages: first, classification is efficient due to dynamic selection (4-class vs 64-class classification). Second, classification errors are confined to neighbors of the true viewpoints. This means a wrongly estimated viewpoint is at most an adjacent viewpoint of the true viewpoint, enabling fast recovery from incorrect estimations. Third, the robust temporal relationship between poses is used to resolve the left-right ambiguities of human silhouettes.

Experiments conducted on both fronto-parallel videos and aerial videos confirm that the solution can achieve accurate pose and trajectory estimation for these different kinds of videos. For example, the “walking on an 8-shaped path” data set (1,652 frames) can achieve the following estimation accuracies: 85 percent for viewpoints and 98.14 percent for poses.

Dynamically tracking the target by unmanned ground vehicles (UGVs) plays a critical role in mobile drone recovery. This study aims to solve this challenge under diverse random disturbances, proposing a dynamic target tracking framework for UGVs based on target state estimation, trajectory prediction, and UGV control.

To mitigate the adverse effects of noise contamination in target detection, the authors use the extended Kalman filter (EKF) to improve the accuracy of locating unmanned aerial vehicles (UAVs). Furthermore, a robust motion prediction algorithm based on polynomial fitting is developed to reduce the impact of trajectory jitter caused by crosswinds, enhancing the stability of drone trajectory prediction. Regarding UGV control, a dynamic vehicle model featuring independent front and rear wheel steering is derived. Additionally, a linear time-varying model predictive control algorithm is proposed to minimize tracking errors for the UGV.

To validate the feasibility of the framework, the algorithms were deployed on the designed UGV. Experimental results demonstrate the effectiveness of the proposed dynamic tracking algorithm of UGV under random disturbances.

This paper proposes a tracking framework of UGV based on target state estimation, trajectory prediction and UGV predictive control, enabling the system to achieve dynamic tracking to the UAV under multiple disturbance conditions.

Positioning is a key task in most field robotics applications but can be very challenging in GPS‐denied or high‐slip environments. The purpose of this paper is to describe a visual odometry strategy using only one camera in country roads.

This monocular odometery system uses as input only those images provided by a single camera mounted on the roof of the vehicle and the framework is composed of three main parts: image motion estimation, ego‐motion computation and visual odometry. The image motion is estimated based on a hyper‐complex wavelet phase‐derived optical flow field. The ego‐motion of the vehicle is computed by a blocked RANdom SAmple Consensus algorithm and a maximum likelihood estimator based on a 4‐degrees of freedom motion model. These as instantaneous ego‐motion measurements are used to update the vehicle trajectory according to a dead‐reckoning model and unscented Kalman filter.

The authors' proposed framework and algorithms are validated on videos from a real automotive platform. Furthermore, the recovered trajectory is superimposed onto a digital map, and the localization results from this method are compared to the ground truth measured with a GPS/INS joint system. These experimental results indicate that the framework and the algorithms are effective.

The effective framework and algorithms for visual odometry using only one camera in country roads are introduced in this paper.

The purpose of this work is to present a procedure for determining the wind drift factor through two-dimensional computational fluid dynamics (CFD) simulations of the wind acting on a wavy sea surface, such that the subjectivity of its estimation is reduced.

The wind drift factor was determined by two-dimensional CFD analyses with open-channel condition. The characteristic wave was determined by the Sverdrup–Munk–Bretschneider (SMB) method. The uncertainty analysis is based on convergence studies using a single parameter refinement (grid and time step).

This procedure allows the estimation of the wind drift factor in a fetch-limited domain. The domain's value in the analyzed region is 0.0519 ± 4.92% which is consistent with the upper values of the wind drift factors reported in the literature.

The use of a three-dimensional domain was impractical with the available computational resources because of the fine mesh required for wave modeling. The uncertainty analysis consisted only of a verification procedure. Validation against real data was not possible because of the lack of measured data in the analyzed region.

The wind drift factor is usually estimated based on either experience or random sampling. The original contribution of this work is the presentation of a CFD procedure for estimating the wind drift factor, in which the domain inlet is subjected to a wave boundary condition and to a wind velocity.

Admittance control is a typical complaint control methodology. Traditionally, admittance control systems are based on a dynamical relationship described by Voigt model. By contrast, after changing connection of spring and damper, Maxwell model produces different dynamics and has shown better impact absorption performance. This paper aims to design a novel compliant control method based on Maxwell model and implement it in a robot catching scenario.

To achieve this goal, this paper proposed a Maxwell model based admittance control scheme. Considering several motion stages involved in one catching attempt, the following approaches are adopted. First, Kalman filter is used to process the position data stream acquired from motion capture system and predict the subsequent object flying trajectory. Then, a linear segments with parabolic blends reaching motion is generated to achieve time-optimal movement under kinematic and joint inherent constraints. After robot reached the desired catching point, the proposed Maxwell model based admittance controller performs such as a cushion to moderate the impact between robot end-effector and flying object.

This paper has experimentally demonstrated the feasibility and effectiveness of the proposed method. Compared with typical Voigt model based compliant catching, less object bounding away from end-effector happens and the success rate of catching has been improved.

The authors proposed a novel Maxwell model based admittance control method and demonstrated its effectiveness in a robot catching scenario. The author’s approach may inspire other related researchers and has great potential of practical usage in a widespread of robot applications.

This paper presents the design and validation of a universal 6D seam tracking system that reduces the need of accurate robot trajectory programming and geometrical databases in robotic laser scanning. The 6D seam tracking system was developed in the flexible unified simulation environment, integrating software prototyping with mechanical virtual prototyping, based on physical experiments. The validation experiments showed that this system was both robust and reliable and should be able to manage a radius of curvature less than 200 mm. In the pre‐scanning mode, a radius of curvature down to 2 mm was managed for pipe intersections at 3 scans/mm, using a laser scanner with an accuracy of 0.015 mm.

The purpose of this paper is to present a 3D model of deep welding of dissimilar metals and to show how to model the electron beam deflection due to thermoelectric fields caused by temperature gradients in some dissimilar metals (Seebeck effect).

A 3D thermoelectric and heat conduction model is used to estimate the deflection of the electron beam used during welding of dissimilar metals. A weak coupling between analysed fields is assembled. Additionally, the influence of the deflection on the calculated fields was not taken into account. The problem is solved using a finite element method.

It is possible to model Seebeck effect in a relative simple way using the finite element approach.

The paper presents a detailed description of modelling procedure of a complex coupled field problem.

For autonomous vehicles, trajectory prediction of surrounding vehicles is beneficial to improving the situational awareness of dynamic and stochastic traffic environments, which is a crucial and indispensable element to realize highly automated driving.

In this paper, the overall framework consists of two parts: first, a novel driver characteristic and intention estimation (DCIE) model is built to indicate the higher-level information of the vehicle using its low-level motion variables; then, according to the estimation results of the DCIE model, a classified Gaussian process model is established for probabilistic vehicle trajectory prediction under different motion patterns.

The whole method is later applied and analyzed in the highway lane-change scenarios with the parameters of models learned from the public naturalistic driving data set. Compared with other traditional methods, the performance of this proposed approach is proved superior, demonstrated by the higher accuracy in the long prediction horizon and a more reasonable description of uncertainty.

This hierarchical approach is proposed to make trajectory prediction accurately both in the short term and long term, which can also deal with the uncertainties caused by the perception system or indeterminate vehicle behaviors.

The purpose of this paper is to focus on perfect trajectory tracking control of 2 DOF non‐holonomic mobile robots in which the guidance and control commands are imposed through independent driver wheels. Model‐based nonlinear controllers for these robots with unknown parameters require estimation of a specified set of the robot parameters. The effects of the proposed model dynamics in both local and global coordinate systems are fully examined on the parameter estimation and tracking performance.

Design of suitable feedback linearization (FL) controllers for trajectory tracking control of wheeled mobile robots (WMRs) is first reviewed. A FL controller whose parameters are tuned using fuzzy computations (fuzzy if‐then rules) is then developed. In the line of the other contributions of the paper, a pure fuzzy controller that is merely based on fuzzy if‐then rules is proposed to trajectory tracking control of the mobile robots.

Use of global dynamics for designing a suitable FL control system leads to a perfect compensation for initial off‐tracks. Furthermore, the estimated parameters are unbiased because the corresponding regressor/signal matrix indicates a high rank of persistent excitation. Fuzzy tuning of the controller instead of keeping the gains fixed makes the overall system more robust against measurement noises while upper bounds and fluctuations of the input torques are both remarkably reduced. The pure fuzzy controller is naturally independent of the robot dynamics and therefore, the necessity of parameter estimation algorithm is removed.

The paper provides some new nonlinear controllers for WMRs, in order to make perfect trajectory tracking and initial off‐tracks compensation.