Xu Shen

Bio: Xu Shen is an academic researcher from University of California, Berkeley. The author has contributed to research in topics: Engineering & Collision avoidance. The author has an hindex of 4, co-authored 10 publications receiving 44 citations. Previous affiliations of Xu Shen include Tsinghua University.

A smooth and undistorted toolpath interpolation method for 5-DoF parallel kinematic machines

Abstract: Comparing with classical serial machining center, Parallel Kinematic Machines (PKMs) are displaying superiority in coupling rotational motion control, for which the tilt-and-torsion (T&T) angle was developed as a powerful research tool. However, this rotation modality is not widely used in manufacturing scenarios due to the lack of associated interpolation algorithm, which therefore impedes the application of PKMs. In this work, a modified spline method is designed to interpolate 5-axis PKM toolpath described by T&T angle. Quintic splines are constructed in both positional and orientational trajectories for smoothness requirement, and the interpolation quality is further reinforced by anti-distortion strategy and coordinating reparameterization spline. Simulation based on actual toolpath and machine model prove that the proposed algorithm realizes the smoothness of acceleration in all toolpath components, decreases the shock to the system, and reduces the surface error after cutting. The results indicate that the interpolation method in this paper can provide new algorithm for PKMs with T&T angle and promote their development in the industrial manufacturing and production.

Autonomous Parking of Vehicle Fleet in Tight Environments.

Abstract: The problem of autonomous parking of vehicle fleets is addressed in this paper. We present a system-level modeling and control framework which allows investigating different vehicle parking strategies while taking into account path planning and collision avoidance. The proposed approach decouples the problem into a centralized parking spot allocation and path generation, and a decentralized collision avoidance control. This paper presents the hierarchical framework and algorithmic details. Extensive simulations are used to assess several allocation strategies in terms of total fleet parking time and queue length. In particular, we observe that when parking large vehicle fleets, a phenomenon similar to Braess's paradox occurs.

Parameter Optimization for the Driving System of a 5 Degrees-of-Freedom Parallel Machining Robot With Planar Kinematic Chains

Abstract: Driving system parameter optimization (DSPO) is an important approach to improve robots' dynamic performances such as acceleration capacity, load carrying capacity, and operation stability. To achieve better dynamic performance, motors with high power and high cost are generally used. But this leads to a waste of resources. It is difficult to simultaneously make the robots satisfy the prescribed requirements and avoid over conservative design. This issue is much more challenging for parallel machining robots due to the coupling characteristics of the closed kinematic chains. In this paper, a 5 degrees-of-freedom (DoF) parallel machining robot with planar kinematic chains is presented, and its dynamic model is established based on the virtual work principle. Then, a DSPO method for 5-DoF machining robots is proposed by considering the classical machining trajectories that can reflect the robots' performance requirements. The motor output under these trajectories and candidate motor parameters are presented in a comprehensive graph. Combined with motor selection criteria, the feasible motors and usable reduction ratio range are derived. To optimize the reduction ratio, a dynamic index is proposed based on the variance degree of the motor output torque to evaluate driving system's operational stability. On this basis, the optimal reduction ratio is obtained by minimizing this index to improve the stability of machining robots. Based on the proposed method, the DSPO for the 5-DoF parallel machining robot is implemented, and the optimal driving units are generated. The proposed method can be used for the DSPO of other 5-DoF parallel machining robots.

Collision Avoidance in Tightly-Constrained Environments without Coordination: a Hierarchical Control Approach.

Abstract: We present a hierarchical control approach for maneuvering an autonomous vehicle (AV) in tightly-constrained environments where other moving AVs and/or human driven vehicles are present. A two-level hierarchy is proposed: a high-level data-driven strategy predictor and a lower-level model-based feedback controller. The strategy predictor maps an encoding of a dynamic environment to a set of high-level strategies via a neural network. Depending on the selected strategy, a set of time-varying hyperplanes in the AV's position space is generated online and the corresponding halfspace constraints are included in a lower-level model-based receding horizon controller. These strategy-dependent constraints drive the vehicle towards areas where it is likely to remain feasible. Moreover, the predicted strategy also informs switching between a discrete set of policies, which allows for more conservative behavior when prediction confidence is low. We demonstrate the effectiveness of the proposed data-driven hierarchical control framework in a two-car collision avoidance scenario through simulations and experiments on a 1/10 scale autonomous car platform where the strategy-guided approach outperforms a model predictive control baseline in both cases.

Trajectory smoothing method using reinforcement learning for computer numerical control machine tools

Abstract: Tool-path codes output by computer-aided manufacturing software for high-speed machining are composed of discontinuous G01 line segments. The discontinuity of these tool movements causes computer numerical control (CNC) inefficiency. To achieve high-speed continuous motion, corner smoothing algorithms based on pre-planning methods are widely used. However, it is difficult to optimize smoothing trajectories in real-time systems. To obtain smooth trajectories efficiently, this paper proposes a neural network-based direct trajectory smoothing method. An intelligent neural network agent outputs servo commands directly based on the current tool path and running state in every cycle. To achieve direct control, motion feature and reward models were built, and reinforcement learning was used to train the neural network parameters without additional experimental data. The proposed method provides higher cutting efficiency than the local and global smoothing algorithms. Given its simple structure and low computational demands, it can easily be applied to real-time CNC systems.

Global G3 continuity toolpath smoothing for a 5-DoF machining robot with parallel kinematics

Abstract: Toolpath smoothing is an important approach to improve robots’ operational stability and machining quality. Nowadays, the corner rounding smoothing and curve fitting smoothing algorithms are usually adopted to process the linear toolpath segments to improve its continuity. But the high order continuity between the fitted curve and its adjacent curves is difficult to be guaranteed. For parallel machining robots (PMRs), the tangential, curvature and curvature derivative discontinuities at the junction may lead to the self-excited vibration of mechanical structure, consequently the machining efficiency and quality are decreased. Under this consideration, a global G3 continuity toolpath smoothing method for five degrees of freedom (5-DoF) PMRs is proposed. The linear segments toolpath generated by the Computer-Aided Manufacturing (CAM) system is first divided into long linear segments (LLSs) and short linear segments groups (SLSGs) through breakpoint searching. At the junction point, a B-spline transition curve is inserted to blend adjacent toolpaths. For the SLSG, the quintic B-spline is adopted to fit the discrete data points, constraint equations about the derivatives at the start and end points are established to achieve G3 continuity with the adjacent transition curves. Based on the proposed method, the smoothing for two test toolpaths is carried out, and experiments on a 5-DoF PMR are conducted to show the validity of the method in motion smoothness.

Kinematic calibration of a 5-axis parallel machining robot based on dimensionless error mapping matrix

Abstract: Accuracy problem is one of the most challenging issues for the application of parallel robots in manufacturing industry, and kinematic calibration is a feasible approach to solve it Although lots of researches have brought up a diversity of calibration methods, there are still rooms for the improvement of the accuracy, efficiency and robustness of these calibration effects In this paper, an improved method for kinematic calibration of a 5-axis parallel machining robot is proposed, which includes a new forward kinematic solution (FKS) based on dual quaternion and a modified error modeling process leading to dimensionless error mapping matrixes (EMMs) On this basis, an iterative identification procedure is schemed, and the kinematics and identification simulations are carried out The kinematics simulation results show that the proposed FKS has wider convergence range and faster computation speed than Levenberg-Marquardt algorithm, while the identification simulation results show that the residual pose errors with the proposed dimensionless EMMs are lower than that with the conventional EMM in various units Additionally, the procedure of the full pose measurement with a laser tracker and an auxiliary tool is introduced, and thereby the contrast experiments of kinematic calibration on the prototype are conducted The experiment results indicate that the residual position and orientation errors based on the dimensionless EMM decrease by 9767% and 8685% of the original values, respectively, at least, and by 7677% and 3865% of that based on the conventional EMM, respectively, at most Consequently, it is further confirmed that the proposed calibration method is effective in enhancing the identification accuracy of the geometric errors and improving the positioning accuracy of the studied parallel robot

Space corner smoothing of CNC machine tools through developing 3D general clothoid

Abstract: Tool paths defined by G01/G02/G03 commands need to be smoothed to eliminate the discontinuities of velocity, acceleration and jerk at the junction points. Because of inherent problems of curve fillets, traditional corner smoothing strategies are limited to the smoothing of corners in plane. This article presents a method to smooth the space corners through blending the position, tangent, curvature and sharpness of the adjacent trajectory segments based on 3D general clothoid splines, which are analytically developed by proposing 3D clothoid with G3 continuity. The 3D general clothoid realizes extending the traditional clothoid from 2-dimension to 3-dimension, and reserves some good properties of the traditional clothoid, i.e. the curve length parameterization and the analytically expressed curvature. It can also achieve higher degree of continuity compared to the traditional 2D clothoid. Based on the proposed 3D general clothoid, a corner smoothing algorithm, which is suitable for the smoothing of any planar or space line-line, line-arc or arc-arc pairs with G3 continuity, is proposed. At the same time, a smoothing-error-constraining-algorithm is developed to constrain the smoothing error within the tolerance. Simulation and experimental results, which are obtained from the smoothing of trajectories containing space corners, trajectories containing planar corners with G02/G03 commands, and trajectories containing planar corners with only G01 commands, demonstrate that the proposed corner smoothing algorithm can effectively increase the machining quality and efficiency.