Design and Evaluation of a Hair Combing System Using a General-Purpose Robotic Arm

Nathaniel Dennler, Eura Shin, Maja J. Mataric, Stefanos Nikolaidis

Abstract

This work introduces an approach for automatic hair combing by a lightweight robot. For people living with limited mobility, dexterity, or chronic fatigue, combing hair is often a difficult task that negatively impacts personal routines. We propose a modular system for enabling general robot manipulators to assist with a hair-combing task. The system consists of three main components: a segmentation module, a path planning module, and a trajectory generation module. We quantitatively evaluate the effectiveness of the paths planned by the system with 48 users and qualitatively evaluate the system with 30 users watching videos of the robot performing a hair-combing task in the physical world. The system is shown to effectively comb different hairstyles.

Introduction

For millions of individuals living with limited mobility or dexterity, simple self-care tasks like combing hair can be exhausting or impossible. While assistive robots exist for tasks like eating or dressing, hair care has remained under-explored. This challenge is twofold: perception and planning. Hair is notoriously difficult for computer vision to track due to its high variability in color and texture. Furthermore, "correct" combing is not merely a geometric problem; it requires following the natural grain and curvature of the hairstyle to avoid tangles and discomfort.

Methodology

The authors propose a three-stage modular pipeline to enable general-purpose robotic arms to perform hair combing.

Segmentation Module

Using a SqueezeNet-based encoder-decoder architecture, the system identifies hair in RGB images. To handle the visual noise of individual strands, the system applies temporal smoothing (averaging masks over time) and spatial filtering to ensure the robot focuses on the most relevant hair volume.

The hair segmentation pipeline, showing the transition from raw image to refined segmentation mask.

Path Planning Module

Instead of treating hair as a static object, the system models it as a differential orientation field. It utilizes Coherence-Enhancing Shock Filters to reduce visual noise and highlight the dominant "flow" of the hair. A Gradient Structure Tensor then calculates the local orientation of the hair at every pixel. When a user selects a starting point, the system "flows" along these orientations to generate a natural stroke, maintaining user agency while reducing physical effort.

Trajectory Generation

The 2D image path is mapped to 3D space using depth data from a Kinect sensor. The robot, a Kinova Gen2, aligns the comb teeth orthogonally to the hair surface and moves along the planned path.

Evaluation and Results

The system was trained on the Figaro 1k dataset and evaluated using 30 hairstyles from the USC Hair Salon Database, covering various lengths and colors.

The robotic system performing combing tasks on various hairstyles, including short, long, straight, and wavy hair.

Path Quality and Functional Success

In a study with 48 participants, the image-based "flow" method was rated significantly higher for both "completeness" and "effectiveness" compared to a baseline mesh-based method. A qualitative study with 30 users confirmed that the system successfully made messy wigs look straighter and neater.

Safety and Perception

Physical testing showed that peak forces stayed below 15N, comparable to other delicate robotic tasks like shaving. Interestingly, users often perceived the robot as using too much force if the mannequin's head moved, highlighting that visual cues of gentleness are as critical as actual force limits in assistive robotics.

Discussion and Future Work

While the system effectively handled various lengths, users noted that the robot's silence was unsettling, suggesting a need for auditory cues. Future work will focus on integrating hybrid position-force control to allow the robot to "feel" tangles and adjust its speed accordingly. Additionally, the system must be validated on more diverse hair types, such as very curly or sparse hair, and optimized for speed to better fit into a user's daily routine.

Summary updated

Abstract

This work introduces an approach for automatic hair combing by a lightweight robot. For people living with limited mobility, dexterity, or chronic fatigue, combing hair is often a difficult task that negatively impacts personal routines. We propose a modular system for enabling general robot manipulators to assist with a hair-combing task. The system consists of three main components. The first component is the segmentation module, which segments the location of hair in space. The second component is the path planning module that proposes automatically-generated paths through hair based on user input. The final component creates a trajectory for the robot to execute. We quantitatively evaluate the effectiveness of the paths planned by the system with 48 users and qualitatively evaluate the system with 30 users watching videos of the robot performing a hair-combing task in the physical world. The system is shown to effectively comb different hairstyles.
BibTeX
@inproceedings{dennler2021design,
  author = {Nathaniel Dennler and Eura Shin and Maja J. Mataric and Stefanos Nikolaidis},
  title = {{Design and Evaluation of a Hair Combing System Using a General-Purpose Robotic Arm}},
  booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems},
  year = {2021},
  url = {https://openreview.net/forum?id=4OHxyklRvy},
}
Figure