INVESTIGATION OF A SOFT TACTILE SENSING SYSTEM FOR HUMAN COLLABORATIVE ROBOTS

Abstract

In recent years, the soft tactile sensing system for a human collaborative robot (HuBot) has received great attention. Various soft tactile sensors along with a soft sensing system (a network of soft sensors) have been developed in literature. The use of carbon nanomaterial/polymer composites (CNPCs) for the soft tactile sensors is promising. However, there is a lack of a systematic approach to designing the CNPC tactile sensing system; besides, a single CNPC tactile sensor warrants a comprehensive study.

In this dissertation, soft tactile sensors made from CNPCs are studied first. Specifically, the distribution method for fabricating the multi-wall carbon nanotube (MWCNT) filled epoxy aliphatic acrylate (EAA)/Aliphatic urethane diacrylate (AUD) composite tactile sensor is studied. The coating method for the MWCNT coated polydimethylsiloxane (PDMS) tactile senor is also studied. Axiomatic Design Theory and Adaptable Design Theory are employed to develop this systematic design approach. To demonstrate the effectiveness of the proposed design approach, a prototype of the soft sensing system for HuBot is developed and tested.

The dissertation led to the following conclusions: (1) It is possible to make MWCNT or single-wall carbon nanotube (SWCNT) filled EAA/AUD tactile sensor with the redox method. (2) The coating method is better than the distribution method in terms of constructing a CNPC tactile sensing system for HuBot. (3) A systematic design approach for the CNPC tactile sensing system for HuBots is effective. (4) A prototype of the adaptable CNPC tactile sensing system is constructed with a preliminary success.

This dissertation has made a few contributions in the field of soft tactile sensors and soft sensor networks. Contribution 1: the new knowledge has been generated regarding the feasibility for the distribution of carbon nanomaterials into the EAA/AUD polymer. Contribution 2: the coating method for CNPC tactile sensors has been found better than the coating method for HuBots. Contribution 3: the systematic design approach to soft CNPC tactile sensor networks for HuBots has been developed. Contribution 4: a prototype of the soft CNPC tactile sensor network has been constructed.