The path towards contact-based physical human-robot interaction

TL;DR

This survey maps the early-stage field of contact-based pHRI, arguing that safety must be woven into perception, planning, and control from design to deployment. It aggregates developments across sensing, modeling, planning, and learning—highlighting data-driven methods (RL, LfD, and DL) as key to handling human intent and uncertainty in close-proximity interactions. The authors emphasize ergonomic, social, and ethical dimensions alongside technical challenges, calling for integrated advances and responsible deployment in workplaces, clinics, and everyday life. By outlining architecture choices and computational needs, the paper provides a blueprint for advancing safe, adaptable, and trustworthy pHRI systems.

Abstract

With the advancements in human-robot interaction (HRI), robots are now capable of operating in close proximity and engaging in physical interactions with humans (pHRI). Likewise, contact-based pHRI is becoming increasingly common as robots are equipped with a range of sensors to perceive human motions. Despite the presence of surveys exploring various aspects of HRI and pHRI, there is presently a gap in comprehensive studies that collect, organize and relate developments across all aspects of contact-based pHRI. It has become challenging to gain a comprehensive understanding of the current state of the field, thoroughly analyze the aspects that have been covered, and identify areas needing further attention. Hence, the present survey. While it includes key developments in pHRI, a particular focus is placed on contact-based interaction, which has numerous applications in industrial, rehabilitation and medical robotics. Across the literature, a common denominator is the importance to establish a safe, compliant and human intention-oriented interaction. This endeavour encompasses aspects of perception, planning and control, and how they work together to enhance safety and reliability. Notably, the survey highlights the application of data-driven techniques: backed by a growing body of literature demonstrating their effectiveness, approaches like reinforcement learning and learning from demonstration have become key to improving robot perception and decision-making within complex and uncertain pHRI scenarios. As the field is yet in its early stage, these observations may help guide future developments and steer research towards the responsible integration of physically interactive robots into workplaces, public spaces, and elements of private life.

Figures (5)

• Figure 1: Classifying HRI based on two factors. a) the degree of independence between human and robot actions, including co-existence: humans and robots occupy separate workspaces without any interference, co-operation: humans and robots share a workspace while working on their respective tasks, and collaboration: humans and robots share a workspace and simultaneously work together on a common task. b) the degree of physical and psychological engagement in the interaction. The four quadrants mark the division between social, physical, social-physical HRI and teleoperation. Each quadrant encompasses various forms of interaction along the co-existence-co-operation-collaboration spectrum.
• Figure 2: Three different scenarios in pHRI: a) Dancing, leveraging direct physical contact, b) Handover task, where the robot transfers an object to the human, c) Co-manipulation, where the human and robot collaborate to accomplish a task through direct contact. Picture reproduced from 10144527 with permission.
• Figure 3: Sensors commonly employed for pHRI: a) F/T sensor mounted on the wrist of the REEEM-C robot, b) RGB-D sensor mounted on the REEM-C head, c) Laser range sensor embedded in each foot of the REEM-C, d) Tactile sensors on the NAO robot: capacitive sensors at the head and hands; on/off bumpers on each foot, e) Flexible resistive pressure sensors mounted on a glove, f) RGB-D sensor mounted on the MOVO robot.
• Figure 4: Human body keypoint detection using RGB and depth sensors. Reproduced with permission from 10000133.
• Figure 5: Schematic illustration of a typical pHRI framework, demonstrating the interconnections between various modules. Emphasizing safety as a critical aspect, the figure highlights the overarching need to address safety at each step, as well as to consider ethics from the start.