Haptic object recognition for multi-fingered robot hands

Abstract

In this paper, we present an approach for haptic object recognition and its evaluation on multi-fingered robot hands. The recognition approach is based on extracting key features of tactile and kinesthetic data from multiple palpations using a clustering algorithm. A multi-sensory object representation is built by fusion of tactile and kinesthetic features.We evaluated our approach on three robot hands and compared the recognition performance using object sets consisting of daily household objects. Experimental results using the five-fingered hand of the humanoid robot ARMAR, the three-fingered Schunk Dexterous Hand 2 and a parallel Gripper are performed. The results show that the proposed approach generalizes to different robot hands.

Haptic identification of objects using a modular soft robotic gripper

“…Previous studies on haptic recognition of objects focus on hands with rigid links [23]- [27]. Paolini et al [28] present a method which uses proprioception to identify the pose of an object in a rigid hand after a grasp.…”

Face classification using touch with a humanoid robot hand

“…Moreover, once the learning completed, in order to minimize the required number of grasp actions, a new grasp location is chosen on the object by maximizing the expected information gain on the probabilistic belief about the type of object. A hybrid approach uses both tactile and haptic information in [6] by completely modelling objects with 3-D point clouds representations, systematically scanning the objects with grippers equipped with tactile sensor arrays, or by creating Self-Organizing Maps from few grasps with more complex robotic hands [7], [8]. In contrast to the methods reviewed previously, it might be preferable to continuously probe an object surface instead of discretely grasp or touch an object: humans do not release and grasp several times an object in order to recognize it by touch but rather follow the surface with their fingers.…”

Recent progress on tactile object recognition

“…The author used a fixed Kinect camera to capture the object in order to detect the deformability, and then the tactile signals were introduced to analyze the internal state. However, this 44,45,46 method requires the 3-D model of the object and therefore it is difficult to be extended to unfamiliar objects. In Figure 9, we list some representative data collection scenes or the algorithm architectures, which are adopted in the corresponding references.…”

“…However, many manipulators also have tactile sensors on their palm, which inspire some scholars to grasp objects with a power grasp. For example, Navarro et al 44 and Schmitz et al 45 used this grasp to recognize multiple objects based on a single layer neural network and deep learning technology. Ma et al 46 also used this power grasp strategy to classify 16 different objects.…”