In Pursuit of Safety: An Open-Source Library for Physical Human-Robot Interaction

Abstract

This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination. 3 IEEE ROBOTICS & AUTOMATION MAGAZINE • JUNE 2018 impact with the operator. Ideally, both solutions should be combined to provide the highest level of safety. Although pHRI has been extensively investigated by the research community, to the best of our knowledge, no general open-source software solution exists to date. Thus, each research team or industrial organization is forced to develop its own applications, limiting the adoption, benchmarking, and growth of pHRI in the community. So the main motivation behind OpenPHRI is to provide a full-featured, opensource software library that can also be easily extended to develop pHRI applications. Overview of the Library The controller, constraints, and inputs described in this article are all available in the OpenPHRI software library, distributed online [14] free of charge under the GNU Lesser General Public License version 3 (LGPLv3) [15]. This license allows integration with open-or closed-source software as long as any modifications made to the library are shared with the community. OpenPHRI is written in C++ to maximize efficiency in terms of computation and memory footprint and to easily embed it in existing projects. Python bindings are also provided, because this language is largely used in the robotics community and because it allows quick prototyping, as most computations are performed in machine language to keep computational times small. An interface with the robotics simulator Virtual Robot Experimentation Platform (V-REP) [16] is also furnished. The V-REP remote Application Programming Interface (API) library embedded in OpenPHRI has no particular license, so it does not violate the LGPLv3. A wrapper for the Robot Operating System framework will be released in the near future. Users can easily integrate other simulators, frameworks, and robots at will. The detailed hierarchy of the project is given in Table 1.

A Deep Learning Framework for Recognizing Both Static and Dynamic Gestures

“…The experiments were performed on BAZAR robot [ 63 ] which has two Kuka LWR 4+ arms with two Shadow Dexterous Hands attached at the end-effectors. We exploited OpenPHRI [ 64 ], which is an open-source library, to control the robot while corroborating safety of the human operator. A finite state machine is developed to control behavior of the robot which is determined by the sensory information for example, hand gestures, distance of the human operator from the robot, joint-torque sensing and so forth.…”

A real-time human-robot interaction framework with robust background invariant hand gesture detection

“…Moreover, the backgrounds are now replaced with random pattern/indoor-architecture images to make the detection robust and background invariant. The vision pipeline is then integrated with OpenPHRI [12] to complement the library with two collaborative modes of the ISO/TS 15066 safety standards. This integration is detailed in Section 6.…”

“…We integrate the proposed hand gesture detection module with our physical human-robot interaction library OpenPHRI [12] for robot control. On one hand, this ensures safety of the human coworker -by complementing the standard collaborative modes in OpenPHRI -while on the other provides a natural means for robot programming and reprogramming, through hand gestures.…”

A collaborative robot for the factory of the future: BAZAR

“…Besides, this simplicity facilitates the adoption of OpenPHRI by a large panel of people. More details about the method and experimental results are presented in [46]. The OpenPHRI library has been validated multiple times on BAZAR.…”