Franz Dietrich

On contact models for assembly tasks: Experimental investigation beyond the peg-in-hole problem on the example of force-torque maps

Force guided assembly is attractive but the implementation is challenging when uncertainties are present. In these cases, contact models that guide the assembly process are attractive. This article elaborates the usage of static contact models, which map displacements to force-torque vectors of particular contact situations (force-torque map). This model, a map of discrete points, contains force and torque values that correspond to position errors. The inversion of this map, using forces and torques measured from an assembly attempt, yields correction movements in order to accomplish the assembly iteratively. A hypothesis stating the dependency of the models quality on its injectivity is investigated. This aspect is studied thoroughly in so-called redundancy maps, which reveal regions of considerable ambiguity of the model. Experimental results are presented, which validate the hypothesis about the dependency of the convergence of the assembly process on the ambiguity of the initial position. In addition to the peg-in-hole problem, which has become a standard scenario to validate force guided assembly, the scope of this article also covers force guided assembly of more complex parts. Here, the analysis gives evidence to believe that it is unlikely that the implementation convergences acceptably, which is validated by experimental results.

RCA562: Control Architecture for Parallel Kinematic Robots

The design of powerful control that suits multiple types of parallel kinematic robots is extraordinarily challenging. The diversity of parallel kinematics and their optimization that customizes them to specific tasks require highly individualized control functionalities. This work intends to provide principles, methods and tools for the development of such control software. It aims at the time-efficient realization of custom robot controllers that suit particular application domains and use hardware resources as sophisticated as possible. A task-frame formalism for trajectory generation is defined exploiting the full potential of parallel robots. This formalism can be used as a generic programming interface for parallel robots. Design patterns for so-called active connectors, modular motion planning, sensor integration and restricted state machines in token-passing context are discussed with regard to control of parallel robots. RCA562, a control application for parallel robots incorporating this knowledge, serves as an illustrative validation example.

Computationally Efficient Adaption of the Window Size of Discrete Position Differentiators

Digital motion control requires precise and low-noise velocity information. Since this velocity information must be calculated from position encoders in each control cycle, time efficiency of these algorithms is a very important design goal. Additionally, it is required that these algorithms operate over wide ranges of both velocity and acceleration. Model-based feedback observers fulfill these requirements for many applications but in some cases they, for various reasons, cannot be formulated. Various applications have been presented in which adaptive filters are applied to overcome the compromises of fixed-length filters. However, it remains desirable to improve the performance of these methods especially in regions of very low velocity and to lower the computation time. This paper presents a new algorithm for the adaption of the window size of differentiator algorithms and a novel criterion for the velocity estimation. This criterion is used to take the dependence between the systems acceleration and the velocity estimation error into account. The properties are studied in simulations and compared to other differentiation techniques. Additionally, results from experiments with a 6-degree-of-freedom servohydraulic Stewart-Gough platform equipped with the new velocity estimation scheme are presented and compared to the performance of concurrent algorithms.

Detection and Avoidance of Singularities in Parallel Kinematic Machines

In this work a geometrical and a physically based method for the detection of singularities are presented, which provide information about the distance of a given position to singularities. The integration of this singularity detection into a robot controller is presented and validated by experimental results. Based on these results a path planning algorithm that avoids singularities is developed. It uses a particle based randomizing scheme for finding a path within the workspace to which a virtual potential field is applied. The HexaII demonstrator of the Collaborative Research Center 562 serves as a validation platform for this algorithm. The singularity avoidance path planner is integrated into the real-time context of the control application RCA562, from which experimental results are presented.

A parallel kinematic concept targeting at more accurate assembly of aircraft sections

This article is concerned with the kinematics for flexible and precise assembly systems for big and heavy parts, e.g. parts of plane fuselages or wings. The hypothesis of this article is that the common 3- PPP S Tripod (which is a current industrial solution to the problem) may be replaced by a 6-S P S parallel kinematic. In difference to many other parallel kinematic machines, the part which has to be assembled is used itself as the end effector platform in this concept. On the part and on the floor, there is a generic grid of fixation points, which gives the possibility for (re)configuration and adaption to suit particular requirements. This article compares the transmission of drive errors in order to evaluate the transmission of alignment increments. This criterion is used for an optimization that searches for a kinematic configuration that outperforms the Tripod systems. It is shown that there is a considerable number of configurations which are superior to the Tripod system. Based on these insights, it is concluded that there is potential for the improvement of current systems through multi-objective optimization.

A-priori Fisher information of nonlinear state space models for experiment design

This article presents advances in optimal experiment design, which are intended to improve the parameter identification of nonlinear state space models. Instead of using a sequence of samples from one or just a few coherent sequences, the idea of identifying nonlinear dynamic models at distinct points in the state space is considered. In this way, the placement of the experiment points is fully flexible with respect to the set of reachable points. Also, a method for model-based generation of prediction errors is proposed, which is used to compute an a-priori estimate of the sample covariance of the prediction error. This covariance matrix may be used to approximate the Fisher information matrix a-priori. The availability of the Fisher matrix a-priori is a prerequisite for experiment optimization with respect to covariance in the parameter estimates. This work is driven by the problem of parameter identification of hydraulic models. There are methods for hydraulic systems regarding the estimation of parameters from experimental data, but the choice of experiments has not been treated adequately yet. A hydraulic servo system actuating a stewart platform serves as an illustrative example to which the methods above are applied.

Enhancements of force-torque map based assembly applied to parallel robots

This article presents the application and enhancements of assembly strategies using force-torque maps in view of parallel robots. Starting from classical assembly strategies the fundamentals of force-torque maps are briefly discussed. Taking up the core idea, a strategy for assembling objects by parallel robots is proposed, where frictional effects are reduced by an enhanced force and position control. Special attention is paid to acquisition of force-torque maps with respect to parameter sensitivity. Practical issues are pointed out and discussed in view of industrial applications. A quality assessment of force-torque maps based on a redundancy index is introduced hinting at the performance and possibilities of optimization of the assembly strategy.

Flexible forming with hexapods

The work presented in this paper provides the basis for an extension of conventional orbital forming to a new, flexible forming process. A hydraulically actuated hexapod is used to move a swage in order to extrude a work piece towards a desired shape. This paper presents an analytical model of the actuators and the dynamics of the hexapod. This model is used for developing a cascaded position control law, and for developing a simulation environment representing the machine and the work piece in an entire forming process loop. The design of the machine is briefly introduced and results of position control are presented.

Novel form-flexible handling and joining tool for automated preforming

Abstract The production rates of carbon fiber reinforced plastic (CFRP) parts are rising constantly which in turn drives research to bring a higher level of automation to the manufacturing processes of CFRP. Resin transfer molding (RTM), which is seen as a production method for high volumes, has been accelerated to a high degree. However, complex net-shape preforms are necessary for this process, which are widely manually manufactured. To face these challenges a new concept for the manufacturing of carbon fiber preforms with a form-flexible gripping, draping and joining end-effector is presented and discussed. Furthermore, this paper investigates the application of this concept, describes the initial build-up of a demonstrator, focusing on material selection and heating technology, and discusses test results with the prototype. This prototype already validates the feasibility of the proposed concept on the basis of a generic preform geometry. After a summary, this paper discusses future in-depth research concerning the concept and its application in more complex geometries.

Disassembly Planning and Assessment of Automation Potentials for Lithium-Ion Batteries

Traction batteries are composed of various materials that are both economic valuable and environmentally relevant. Being able to recover these materials while preserving its quality is not only economically attractive, but it can also contribute to decrease the environmental impact of electric vehicles. Disassembly can play in this regard a key role. On the one hand it might allow to separate potential hazardous substances and avoid an uncontrolled distribution of these substances into other material flows. One the other hand disassembly might promote improving the rate of material recovered while preserving its quality and decreasing disassembly costs. In this chapter we present a methodology for the estimation of disassembly sequences and for the estimation of automation potentials for the disassembly of traction batteries. The methodology is illustrated with an experimental case study.