Humans are able to adapt their movements according to properties, such as weight, size, shape, or temperature, of the objects we interact with. Robots that collaborate with humans should know how to interpret such implicit signals.
As most previous studies focused on the physical properties of handled objects, a recent paper investigates the carefulness (the caution and attention that humans exercise when handling an object).
Image credit: Cristina Zaragoza/Unsplash, free licence
This property is influenced by multiple factors, like physical characteristics, emotional attachment, or economic value. The researchers propose an online implementation of a classifier that distinguishes between careful and non-careful motions using data from a low-resolution camera.
The experimental results show that the system successfully works online and can generalize over unknown human subjects and new kinds of transportation motions.
When manipulating objects, humans finely adapt their motions to the characteristics of what they are handling. Thus, an attentive observer can foresee hidden properties of the manipulated object, such as its weight, temperature, and even whether it requires special care in manipulation. This study is a step towards endowing a humanoid robot with this last capability. Specifically, we study how a robot can infer online, from vision alone, whether or not the human partner is careful when moving an object. We demonstrated that a humanoid robot could perform this inference with high accuracy (up to 81.3%) even with a low-resolution camera. Only for short movements without obstacles, carefulness recognition was insufficient. The prompt recognition of movement carefulness from observing the partner’s action will allow robots to adapt their actions on the object to show the same degree of care as their human partners.
Research paper: Lastrico, L., Carfì, A., Rea, F., Sciutti, A., and Mastrogiovanni, F., “From Movement Kinematics to Object Properties: Online Recognition of Human Carefulness”, 2021. Link: https://arxiv.org/abs/2109.00460