Soft Finger Grasp Force and Contact State Estimation from Tactile Sensors

Hun Jang,Joonbum Bae,Kevin Haninger

2024-10-26

Abstract:Soft robotic fingers can improve adaptability in grasping and manipulation, compensating for geometric variation in object or environmental contact, but today lack force capacity and fine dexterity. Integrated tactile sensors can provide grasp and task information which can improve dexterity,but should ideally not require object-specific training. The total force vector exerted by a finger provides general information to the internal grasp forces (e.g. for grasp stability) and, when summed over fingers, an estimate of the external force acting on the grasped object (e.g. for task-level control). In this study, we investigate the efficacy of estimating finger force from integrated soft sensors and use it to estimate contact states. We use a neural network for force regression, collecting labelled data with a force/torque sensor and a range of test objects. Subsequently, we apply this model in a plug-in task scenario and demonstrate its validity in estimating contact states.

Robotics

What problem does this paper attempt to address?

The problem that this paper attempts to solve is that the soft - robot fingers lack sufficient force capacity and fine dexterity when grasping and manipulating objects. Specifically, although soft fingers can improve the adaptability of grasping by adapting to the geometric changes of object or environmental contact, they are currently insufficient in force sensing and fine - scale manipulation. In addition, integrating tactile sensors can provide grasping and task information to improve dexterity, but ideally should not require training for specific objects.

To overcome these problems, the authors propose a method for estimating the total force exerted by the fingers based on integrated soft sensors (such as pressure and strain sensors), and use a neural - network regression model for force estimation. Through this method, they hope to accurately estimate the contact state and improve the stability and control accuracy of grasping without specific training for each object. This includes estimating the force of a single finger and the external force exerted on the grasped object by the combined force of multiple fingers, thereby achieving task - level control.

In summary, the main objectives of the paper are:
1. Improve the force - sensing ability and manipulation accuracy of soft fingers in grasping.
2. Estimate the contact state without specific - object training by integrating tactile sensors and machine - learning methods.
3. Verify the effectiveness of the proposed method in practical applications such as plug - in and - out tasks.

These efforts aim to enable soft - robot fingers to handle objects of various shapes and sizes more effectively while maintaining their inherent adaptability and safety.