Johannes Schrimpf

Experiments towards automated sewing with a multi-robot system

In this paper a concept for automated multi-robot-aided sewing is presented. The objective of the work is to demonstrate automatic sewing of 3D-shaped covers for recliners, by assembling two different hide parts with different shapes, using two robots to align the parts during sewing. The system consists of an industrial sewing machine and two real-time controlled Universal Robots 6-axis industrial manipulators. A force feedback system combined with optical edge sensors is evaluated for the control of the sewing process. The force sensors are used to synchronize the velocity and feed rate between the robots and the sewing machine. A test cell was built to determine the feasibility of the force feedback control and velocity synchronization. Experiments are presented which investigate the ability of the robot to feed a hide part into the sewing machine using a force sensor and different strategies for velocity synchronization.

Set-Based Tasks within the Singularity-Robust Multiple Task-Priority Inverse Kinematics Framework: General Formulation, Stability Analysis, and Experimental Results

Inverse kinematics algorithms are commonly used in robotic systems to transform tasks to joint references, and several methods exist to ensure the achievement of several tasks simultaneously. The multiple task-priority inverse kinematics framework allows tasks to be considered in a prioritized order by projecting task velocities through the nullspaces of higher priority tasks. This paper extends this framework to handle setbased tasks, i.e. tasks with a range of valid values, in addition to equality tasks, which have a specific desired value. Examples of set-based tasks are joint limit and obstacle avoidance. The proposed method is proven to ensure asymptotic convergence of the equality task errors and the satisfaction of all high-priority set-based tasks. The practical implementation of the proposed algorithm is discussed, and experimental results are presented where a number of both set-based and equality tasks have been implemented on a 6 degree of freedom UR5 which is an industrial robotic arm from Universal Robots. The experiments validate the theoretical results and confirm the effectiveness of the proposed approach.

Real-time analysis of a multi-robot sewing cell

This paper presents a sewing cell based on an industrial sewing machine, and with two robots controlling the sewing operation in real-time. The software architecture and the communication structure are presented. Widely available software and hardware technologies were used to build a highly flexible system for operation and control prototyping. This paper focuses on analyzing the real-time characteristics of the control system. Experiments were conducted to measure the delays in different parts of the system in order to gain an understanding of the real-time performance and to show that the chosen system is capable to make use of the robots high update frequency and low tracking delay.

Real-time system integration in a multi-robot sewing cell

The sewing process is a manufacturing technology which presents severe challenges for automation. Due to variations in the material properties and unpredictable mechanical compliance, real-time sensor-based control strategies are necessary to achieve satisfying results. This paper presents a sewing cell consisting of two lightweight industrial robots and a sewing machine. Sensors are included both for force and edge positioning control. Experiments are presented showing the performance of the proposed real-time control framework. The experiments focus on the real-time control loop for force and edge control during the sewing process.

Velocity coordination and corner matching in a multi-robot sewing cell.

Automated sewing is a complicated task in manufacturing. Due to the non-rigid work pieces and variations in the material characteristics, sensor-based control has to be used to accomplish the sewing operation. This paper presents a strategy for velocity synchronization and corner matching in an automated sewing cell based on two industrial manipulators and a sewing machine. A hybrid force/motion control scheme is adopted using feedback from force/torque sensors for tension control and optical sensors to control the seam position. The strategy is based on switching between force control and displacement control using a leader/follower coordination scheme. This addresses the problem of corner mismatch occurring when two independent force controllers are used for controlling the two robots. Experiments verify that the proposed method gives a satisfactory corner matching, which is crucial for the presented sewing case.

Experimental results for set-based control within the singularity-robust multiple task-priority inverse kinematics framework

Inverse kinematics algorithms are commonly used in robotic systems to achieve desired behavior, and several methods exist to ensure the achievement of numerous tasks simultaneously. The multiple task-priority inverse kinematics framework allows a consideration of tasks in a prioritized order by projecting task velocities through the null-spaces of higher priority tasks. Recent results have extended this framework from equality tasks to also handling set-based tasks, i.e. tasks that have an interval of valid values. The purpose of this paper is to further investigate and experimentally validate this algorithm and its properties. In particular, this paper presents experimental results where a number of both set-based and equality tasks have been implemented on the 6 Degree of Freedom UR5 which is an industrial robotic arm from Universal Robots. The experiments validate the theoretical results.

Model-based feed-forward and setpoint generation in a multi-robot sewing cell

This paper describes an automated sewing system that focuses on sewing of curved edge segments. The sewing cell includes an industrial sewing machine as well as three industrial robots that handle the parts during the sewing process. Based on experiences from previous work, several improvements are presented that address issues with previous implementations. The main contribution concerning the control system is a model-based generation of the feed-forward velocity for the robot tool movement, based on a geometric description of the parts. This approach increases performance, especially in the case of curved fabric, as well as robustness in case of faulty sensor data. Additionally, this paper describes hardware upgrades that address mechanical challenges that were described in former publications. Finally, the paper presents experiments comparing the model-based control approach with the former control system based on constant feed-forwards velocity and setpoints.

Fast dual-arm manipulation using variable admittance control: Implementation and experimental results

This paper presents a control system for fast cooperative dual-arm manipulation of rigid objects with experimental results. The motivation for multi-arm manipulation comes from the wide range of applications. The possible tasks that can be performed by such a system greatly exceed those of a single manipulator system. The proposed system is flexible with respect to uncertainties in object size. Moreover, it allows for physical human interaction through force/torque sensing. This is especially beneficial in industrial cases where humans and robots work on the same production line. One main goal of this paper is to bridge the gap between current research regarding dual-arm manipulation and the implementation possibilities on current industrial robots and widely available standard hardware.

Differential feed control applied to corner matching in automated sewing

This paper presents a new method for independent feed control in an automated sewing cell. This is important to match the corners of the parts as well as to compensate for uncertain material characteristics and variations in the length of the parts. In this method, the feed speed for the two parts is controlled independently, based on measurements of the endpoints of the parts while keeping an equal sewing force in both parts. Different strategies for correcting errors are presented and experiments are shown to evaluate the different strategies. Possibilities for using the methods to match reference points during the sewing are discussed.

Time-analysis of a real-time sensor-servoing system using line-of-sight path tracking

This paper presents a study of the real-time control conditions for a robotic system with visual servo control. The system is based on an industrial manipulator with a modified controller allowing real-time joint-level control. The work is particularly concerned with delays and path deviations. The focus is on the Line-of-Sight based path tracking controller. The paper describes an analysis of the different delays in the sensor-robot system and a model is presented based on the overall delay. Further the stability of the modeled system is analyzed in respect to a specific control case. The correctness of the estimated system delay is indicated by comparison of simulated and experimental results.