Skip to main content
×
Home
    • Aa
    • Aa

Recognising and manipulating objects using data from a whisker sensor array

  • R. Andrew Russell (a1) and Jaury Adi Wijaya (a1)
Abstract

For many biological creatures sensory whiskers are an effective means of detecting and recognising nearby objects. The project described in this paper has the aim of demonstrating that whisker sensors can be used as a similarly effective form of robot sensing. Many mobile robots have used whiskers as simple switches to warn of an imminent collision. However, these devices cannot provide the detailed surface profile information required to recognise and accurately locate objects. Several research groups have built advanced whisker sensors that can determine the position of a contact along the length of the whisker. Although these whisker sensors are usually little more than a length of flexible spring material they do require complex sensing and actuation mechanisms at the whisker root. In this project an array of eight whisker sensors is scanned over external objects by the motion of the robot. The resulting deflection of the whiskers is monitored by a potentiometer at each whisker root. By recording the deflection of the whiskers as they slide over external objects sequences of surface points can be determined. Object recognition algorithms have been developed that allow the robot to recognise, grasp and retrieve a range of objects using the whisker data. In this paper the robot, WhiskerBOT, is described together with the object recognition and localisation algorithms. Results of practical experiments are also presented.

Copyright
Recommend this journal

Email your librarian or administrator to recommend adding this journal to your organisation's collection.

Robotica
  • ISSN: 0263-5747
  • EISSN: 1469-8668
  • URL: /core/journals/robotica
Please enter your name
Please enter a valid email address
Who would you like to send this to? *
×

Keywords:

Metrics

Full text views

Total number of HTML views: 0
Total number of PDF views: 15 *
Loading metrics...

Abstract views

Total abstract views: 158 *
Loading metrics...

* Views captured on Cambridge Core between September 2016 - 22nd September 2017. This data will be updated every 24 hours.