Currently I am working on safety issues during human robot interactions, i.e. how to design a safe controller for a co-robot (a robot operates in a human-involved environment). This problem can be broken into several sub-problems and I am considering the following ones: (1) multi-agent interactions (2) sensing and knowledge representations (3) learning and predictions (4) human modeling and (5) constrained optimal control and decision-making.
- 2014~ now Robustly-Safe Automated Driving (ROAD) Systems, UC Berkeley
- 2013~ now Robot Safe Interaction Systems (RSIS) for Intelligent Industrial Co-Robots, UC Berkeley
Abstract: With the development of modern robotics, robots are entering people's life in multiple ways. As identified by National Robotics Initiative (NRI), future intelligent robots can be co-defenders, co-explorers, co-inhabitants and even co-workers to human. To successfully launch those co-robots, we must make sure that they are safe to human users. However, this is not a easy task as the robots are operating in a dynamic uncertain environment (DUE) together with other intelligent agents such as humans. In this project, we address the safety issues in the context of (1) multi-agent interactions (2) sensing and knowledge representations (3) learning and predictions (4) human modeling and (5) constrained optimal control and decision-making. (Spotlight Talk in 2015 Bay Area Robotics Symposium.)
- C. Liu, and M. Tomizuka, "Algorithmic safety measures for intelligent industrial co-robots", in Robotics and Automation (ICRA), 2016 IEEE International Conference on.
- C. Liu, and M. Tomizuka, "Safe exploration: addressing various uncertainty levels in human robot interactions", in American Control Conference, 2015.
- C. Liu, and M. Tomizuka, "Control in a safe set: addressing safety in human-robot interactions", in ASME 2014 Dynamic Systems and Control Conference. Best Student Paper Finalist.
- C. Liu, and M. Tomizuka, "Modeling and controller design of cooperative robots in workspace sharing human-robot assembly teams", in Intelligent Robots and Systems (IROS), 2014 IEEE/RSJ International Conference on.
- 2013~ now Contactless human-machine interactions using novel sensors, UC Berkeley
- 2012~2013 Design and control of variable stiffness devices, UC Berkeley
- 2012~2012 Analysis of photoresist sprays in IC Manufacturing, Tsinghua University
- C. Liu, "Analysis of macro and micro properties of ultrasonic spray based on droplet collision and evaporation", Undergraduate Thesis, Tsinghua University, 2012.
Abstract | Publications
Abstract: Vision is the most important perception for human. It is also a good choice for machine perception since it is contactless and provides multi-dimensional information. In this study, I played with two novel sensors: (1) Intel® RealSense™ 3D Camera and (2) Leap Motion. We built a keyboardless typing system using (1), which allows a user to type in the air according his usual typing gesture. We developed an algorithm to detect the typing gesture from the noisy vision input. This algorithm can also learn the user's typing pattern for improved detection. We also built a gesture controlled cart using (1). The cart follows the human user's gestures to turn (wave gesture), accelerate (thumb up) and decelerate (thumb down) in real time. Recently, I built a simulation system using Leap Motion to test the robot's response in the presense of unpredictable human agents. I believe the potentials of these sensors in the robot safety system are huge.
Abstract: Nowadays, most mechanically actuated devices have fixed stiffness as an intrinsic mechanical property. However, this property is undesired when it comes to human robot interactions, as it will cause huge impact if a stiff robot hits a human. Moreover, when we look at how human motions are actuated, we notice that the stiffness of human muscles (considerd as actuators) is changeable. Inspired by this biological fact and motivated for soft and compliant human robot interactions, we designed a variable stiffness actuator (VSA) and made a first prototype to demonstrate the idea. See also this page.
Abstract: In the era of information explosion, the demand for electronic devices is soaring, which puts great pressure on the integrated circuit (IC) production. It is important for manufacturers to increase productivity and lower the cost. During the IC manufacturing process, the photoresist coating is a key step and it will directly affect the photolithography process, thus determine the quality of the circuit. In this project, I studied the coating quality by analyzing the dynamics of photoresist sprays. The photoresist spray is a two phase flow (air and photoresist droplets). The dynamics of the micro photoresist droplets is affected by the airflow parameters and the atomization parameters. Based on simulation and experiment results, I proposed a set of optimal parameters for perfect coating and minimized waste.