Continuous Marker Patterns for Representing Contact Information in Vision-Based Tactile Sensor: Principle, Algorithm, and Verification

摘要

Tactile sensing is an important way for robots to perceive and interact with the environment. Vision-based tactile sensors have been widely developed because of their multimodal measurement function, simple structure, and high resolution. The marker pattern method is the most popular method for vision-based tactile sensors to represent contact information. However, the current mainstream used method, distributed mark pattern (DMP), does not fully exploit the performance advantages and has limitations in terms of precision, resolution and reliability. This article innovatively proposes the idea of continuous marker pattern (CMP) and three basic design principles to improve measurement performance. A specific implementation of CMP and the corresponding information extraction algorithm are introduced in detail. We also develop a simulation process and sensor prototypes to verify the proposed method. Some comparative experiments on measurement performance were carried out to show that the tactile sensors with CMP outperformed DMP regarding measurement precision, resolution, and reliability.