6DoF Robot Assembly Station of Consumer Electronic Production

Assembly of consumer electronics in manufacturing is a time-consuming and labor-intensive task. Assembly is one of the most important robot application in computer, communication, consumer electronic (3C) production. Programing a traditional industrial robotic manufacturing system requires a significant amount of time and resources (and therefore investment). This makes it difficult for production line to switch from one product to the next in a cost-effective way. Because of that high up-front cost, the life cycle of a production line is 2-5 years. However, the accelerated pace of product innovation has reduced the life cycle of each product to 3 - 6 months. The change in production cost comes from 3 aspects: (i) new fixture design, (ii) system fine-tuning, (iii) system re-calibration. To make robots and solutions more competitive in this field, there is a need to develop a workstation that avoid these 3 steps, or minimize human configuration efforts to simplify assembly works. This is required to reduce the effective life cycle of a production line to match the ones of the products. The objective of this project is to develop new manipulation technology that enables automatic assembly of delicate parts onto PCB without the need of expensive re-programming of the robot. The plan is to leverage machine learning to help interpret the information from the various sensors (force feedback, visual sensors, etc.) and train a robotic system to adequately grab various components and insert these components into the pre-designated slots on a PCB.

Abstract:

Policy learning for delicate industrial insertion tasks (e.g., PC board assembly) is challenging. This paper considers two major problems: how to learn a diversified policy (instead of just one average policy) that can efficiently handle different workpieces with minimum amount of training data, and how to handle defects of workpieces during insertion. To address the problems, we propose tolerance-guided policy learning. To encourage transferability of the learned policy to different workpieces, we add a task embedding to the policy’s input space using the insertion tolerance. Then we train the policy using generative adversarial imitation learning with reward shaping (RS-GAIL) on a variety of representative situations. To encourage adaptability of the learned policy to handle defects, we build a probabilistic inference model that can output the best inserting pose based on failed insertions using the tolerance model. The best inserting pose is then used as a reference to the learned policy. This proposed method is validated on a sequence of IC socket insertion tasks in simulation. The results show that 1) RS-GAIL can efficiently learn optimal policies under sparse rewards; 2) the tolerance embedding can enhance the transferability of the learned policy; 3) the probabilistic inference makes the policy robust to defects on the workpieces.

Abstract:

Delicate industrial insertion tasks (e.g., PC board assembly) remain challenging for industrial robots. The challenges include low error tolerance, delicacy of the components, and large task variations with respect to the components to be inserted. To deliver a feasible robotic solution for these insertion tasks, we also need to account for hardware limits of existing robotic systems and minimize the integration effort. This paper proposes a composable framework for efficient integration of a safe insertion policy on existing robotic platforms to accomplish these insertion tasks. The policy has an interpretable modularized design and can be learned efficiently on hardware and transferred to new tasks easily. In particular, the policy includes a safe insertion agent as a baseline policy for insertion, an optimal configurable Cartesian tracker as an interface to robot hardware, a probabilistic inference module to handle component variety and insertion errors, and a safe learning module to optimize the parameters in the aforementioned modules to achieve the best performance on designated hardware. The experiment results on a UR10 robot show that the proposed framework achieves safety (for the delicacy of components), accuracy (for low tolerance), robustness (against perception error and component defection), adaptability and transferability (for task variations), as well as task efficiency during execution plus data and time efficiency during learning.