Safety Assessment Review of a Dressing Assistance Robot

Abstract

Hazard analysis methods such as HAZOP and STPA have proven to be effective methods for assurance of system safety for years. However, the dimensionality and human factors uncertainty of many assistive robotic applications challenges the capability of these methods to provide comprehensive coverage of safety issues from interdisciplinary perspectives in a timely and cost-effective manner. Physically assistive tasks in which a range of dynamic contexts require continuous human–robot physical interaction such as e.g., robot-assisted dressing or sit-to-stand pose a new paradigm for safe design and safety analysis methodology. For these types of tasks, considerations have to be made for a range of dynamic contexts where the robot-assistance requires close and continuous physical contact with users. Current regulations mainly cover industrial collaborative robotics regarding physical human–robot interaction (pHRI) but largely neglects direct and continuous physical human contact. In this paper, we explore limitations of commonly used safety analysis techniques when applied to robot-assisted dressing scenarios. We provide a detailed analysis of the system requirements from the user perspective and consider user-bounded hazards that can compromise safety of this complex pHRI.

Enhancing reliability analysis of AI systems using STPA: a case study on the DroneResponse UAV system

“…In the domain of nuclear power plant operation automation, the STPA method can be applied to the operational automation systems of nuclear power plants, analyzing the potential risks that the application of artificial intelligence brings to the safety of nuclear power plant systems, and proposing corresponding safety control measures [11] . In the field of intelligent robotics, STPA can be used to analyze safety issues in intelligent robot systems, including risks related to robot-human interaction, environment perception, decision-making, and execution [12]- [15] . However, the researchers mentioned above hold a more high-level view and focus more on autonomy of AI systems, neglecting the characteristics of machine learning (ML) components, which is also a direct cause of system safety and reliability.…”

Collision Design

“…Recent robotics advances require physical collaboration between robots and human operators, for example assistive robots at home. In this situation, contact between robot and operator or end-user is inevitable, such as a dressing-support robot [12]; however focus is primarily on keeping this contact 'safe'; i.e. on limiting force involved in contact and avoiding any collisions with humans [45].…”

Evaluation of a Motion Measurement System for PET Imaging Studies

“…Haddadin's works on collision detection, evaluation, and reaction [7] [8], as well as Dogramadzi's methods of safety assessment, especially for robot hazard and error [9], are relevant to the risk assessment and safety design of this type of system. However, this paper focuses on the situation where the robotic system functions normally, and collision is avoided through measurement and compensation of motion, whereas future safety design will focus on ensuring correct system operation, including redundant motion sensing, and evaluating the risk due to failure.…”