Active Bionic Antennas for Object Detection and 3D Localization with Ego-Motion Cancellation

Team

Supervisors

Table of content

  1. Abstract
  2. Related works
  3. Methodology
  4. Experiment Setup and Implementation
  5. Results and Analysis
  6. Conclusion
  7. Publications
  8. Links

Abstract

Inspired by biological systems such as insect antennae and mammalian whiskers, this project focuses on developing bionic tactile sensors for robotic platforms. These sensors enable object detection and 3D localization in unstructured environments. A major innovation is the cancellation of self-induced sensory interference (ego-motion) using Echo State Networks (ESNs). The system improves spatial awareness and mapping by integrating tactile sensing with proprioceptive feedback and adaptive learning.


The literature survey explores bio-inspired tactile sensors, particularly in insects and mammals, highlighting key features such as mechanoreception, active exploration, and adaptation. Early research in robotic whiskers and antennae showed promise but struggled with motion noise. Recent studies introduced ESNs, deep learning, and reinforcement learning for filtering self-sensing artifacts and improving localization in robotics.


Methodology


Experiment Setup and Implementation


Results and Analysis


Conclusion

This project successfully demonstrates the potential of biologically-inspired tactile sensing combined with machine learning for robust robotic navigation. The integration of self-sensing cancellation and tactile-based SLAM opens new possibilities for exploration, disaster response, and industrial applications.


Publications