Robot manipulation in unstructured environments depends on understanding the geometric and physical constraints that the environment imposes on the robot. While approaches like simultaneous localization and mapping (SLAM) allow robots to understand geometric constraints, the understanding of environmental physical manipulation constraints is largely unaddressed when exploring unstructured flexible environments. The aim of this paper is to investigate algorithms that enable robots to autonomously perform exploratory manipulation tasks to comprehend physical constraints that guide safe manipulation. To achieve this goal we break new ground in the area of constraint exploration in unknown flexible environments. We propose using spatial stiffness, represented using screw theory, as a measure of mechanical constraints in the context of flexible environments. This paper focuses on developing a compact representation of global workspace constraints using locally measured stiffness properties. Additionally, we propose methods for identifying and classifying the various types of mechanical constraints that may exist in an elastic workspace, using only local stiffness properties. Used in tandem, these methods form a real-time compatible approach for exploring and mapping constraints of a flexible unknown environment.

Robot-assisted cochlear implant surgery was proposed and proved to be efficient in reducing insertion forces on acrylic scala tympani models. During experiments, the authors discovered that the insertion force not only depends on the shape discrepancy between the scala tympani and the inserted electrode array, but also on the insertion speed. This paper presents a friction model that describes the whole insertion process and investigates the relationship between the insertion speed and the insertion force. Experimental and statistical results show the effectiveness of the model. Applying the friction model generates safety insertion force boundaries for future insertions and gives the optimal insertion speed. It also provides predictive force information for insertion speed feedback control law design which may be applied to robot-assisted cochlear implant surgeries.