| 1. |
- Ji, Qinglei, 1993-
(author)
-
Learning-based Control for 4D Printing and Soft Robotics
- 2022
-
Doctoral thesis (other academic/artistic)abstract
- Exploiting novel sensors and actuators made of flexible and smart materials becomes a new trend in robotics research. The studies on the design, production, and control of the new type of robots motivate the research fields of soft robots and 4D printed robots. 3D Printing (3DP) is an additive manufacturing technology that is widely used in printing flexible materials to fabricate soft robots. 4D Printing (4DP) combines 3DP technologies with smart materials to produce transformable devices. 4DP first prints structures with specifically designed responsive materials. When external stimuli such as temperature, voltage, or magnetic field are applied to the printed structure, it changes shape in a programmable way. The shape morphing property of 4DP makes it a novel approach to the actuators of robots.The employment of these special materials empowers these new robots with better compliance and adaptability to the working environment. However, compared with the rigid counterparts, they also have complex dynamic properties such as substantial non-linearity and time-variance. These factors make the precise modeling and robust control of these new robots challenging and thus hinder their potential applications. Focusing on soft robotic systems enabled by 3DP and 4DP approaches, this dissertation studies both traditional and Machine Learning (ML)-based approaches to the modeling, perception, and control of soft, non-linear, and time-variant robotic systems. The main contributions of this dissertation are:The scheme of Closed-Loop (CL) controlled 4DP (CL4DP) using temperature stimulated Shape Memory Polymer (SMP) is designed and validated numerically and experimentally. The feedback control system increases the precision and robustness of the shape morphing process of 4D printed SMP. Applications of CL4DP are explored.Data-driven model identification methods are applied to learn the dynamic model of the shape morphing process of CL4DP and the learned model has good quality to support model-based control design. Model-free and adaptive Reinforcement Learning (RL) controllers are developed to deal with the non-linearity and time variance of 4D printed actuators. To improve the stability and quick adaptability, a concise basis function set is selected instead of blindly using Deep Neural Networks (DNNs).A quadruped robot enabled by soft actuators and its simulation model are developed. The computation efficiency and model accuracy of the simulator are studied and optimized by comparing different simulation methods such as Finite Element Method (FEM) and lumped parameter method.The optimal walking gait pattern of a soft-legged quadruped robot is found by grid parameter search and RL with a physics based simulation model. To speed up the RL training process, modeling tricks are used to reduce the simulation time of the model and curriculum learning is used to reduce the learning time.A soft sensor made by printable conductive materials and 3DP is designed and optimally calibrated to estimate the shape of a pneumatically driven soft actuator. The geometry of the soft sensor is optimally designed for the best linearity, hysteresis and drift properties. The online estimation is based on a linear regression model learned from experimental data.A pneumatically driven soft gripper is developed by 3DP, the printable soft sensor, and pole-placement control methods. The operation of the gripper does not require an external image feedback system to measure its shape, which is estimated by the integrated soft sensor. The position feedback by the soft sensor and the controller by the pole-placement method enable the soft gripper to perform complex tasks with high precision.
|
|
| 2. |
- Liu, Sichao, 1991-
(author)
-
Multimodal Human-Robot Collaboration in Assembly
- 2022
-
Doctoral thesis (other academic/artistic)abstract
- Human-robot collaboration (HRC) envisioned for factories of the future would require close physical collaboration between humans and robots in safe and shared working environments with enhanced efficiency and flexibility. The PhD study aims for multimodal human-robot collaboration in assembly. For this purpose, various modalities controlled by high-level human commands are adopted to facilitate multimodal robot control in assembly and to support efficient HRC. Voice commands, as a commonly used communication channel, are firstly considered and adopted to control robots. Also, hand gestures work as nonverbal commands that often accompany voice instructions, and are used for robot control, specifically for gripper control in robotic assembly. Algorithms are developed to train and identify the commands so that the voice and hand gesture instructions are associated with valid robot control commands at the controller level. A sensorless haptics modality is developed to allow human operators to haptically control robots without using any external sensors. Within such context, an accurate dynamic model of the robot (within both the pre-sliding and sliding regimes) and an adaptive admittance observer are combined for reliable haptic robot control. In parallel, brainwaves work as an emerging communication modality and are used for adaptive robot control during seamless assembly, especially in noisy environments with unreliable voice recognition or when an operator is occupied with other tasks and unable to make gestures. Deep learning is explored to develop a robust brainwave classification system for high-accuracy robot control, and the brainwaves act as macro commands to trigger pre-defined function blocks that in turn provide micro control for robots in collaborative assembly. Brainwaves offer multimodal support to HRC assembly, as an alternative to haptics, auditory and gesture commands. Next, a multimodal data-driven control approach to HRC assembly assisted by event-driven function blocks is explored to facilitate collaborative assembly and adaptive robot control. The proposed approaches and system design are analysed and validated through experiments of a partial car engine assembly. Finally, conclusions and future directions are given.
|
|
| 3. |
- Danielsson, Oscar, 1982-
(author)
-
Augmented reality smart glasses as assembly operator support : A framework for enabling industrial integration
- 2022
-
Doctoral thesis (other academic/artistic)abstract
- Manufacturing industry is seeing vast improvements in productivity and flexibility as the fourth industrial revolution continues to unfold. However, despite improved computation and automation capacity, there is still a role for operators to play in Industry 4.0, mirrored in the concept of Operator 4.0. Improved productivity and a more competitive global market have contributed to increasing manufacturing complexity, putting greater cognitive demands on operators. A technology that can support operators in this new manufacturing landscape is augmented reality (AR), specifically, headworn AR smart glasses (ARSG). With ARSG, operators can receive information interactively in real time, hands free and overlying their natural environment. ARSG are an emerging technology that is becoming more mature; there are early examples of their use in manufacturing industry, but ARSG are not yet widespread.Because ARSG are an emerging technology, there is still uncertainty as to how ARSG can be integrated, like other production equipment, in assembly lines. When current literature was analyzed, it was found that there is a need for more knowledge particularly from the manufacturing engineering perspective of practically integrating ARSG on the industrial shop floor in the long term. This thesis therefore aims to create a framework that supports industry in making strategic and practical decisions about integrating ARSG in production as an assembly operator support tool. The framework is designed to guide industrial decision makers in evaluating the suitability of ARSG as support in an assembly station and, further to offer specific recommendations and rationales for actions to take. It has two main perspectives: the operators using the ARSG and the manufacturing engineers conducting the integration into the production systems. The framework was iteratively developed, using design science combining qualitative and quantitative methods into mixed methods. Three research questions were developed and answered as steps toward creating and evaluating the framework.The results of the thesis show that ARSG integration should be considered in relation to the investment cost and efficiency gains. For instance, ARSG requires the digitalization of assembly instructions before it can be feasible. If operators are mostly stationary when working and have little need for spatial guidance, there might be cheaper alternatives to ARSG, such as monitors or pick-by-light, that merit prior consideration. The framework has been developed and tested iteratively with industrial experts from different fields, with the initial strawman design based on three literature reviews and previous research.
|
|
