Abstract
The popularity of mobile robots in factories, warehouses, and hospitals has raised safety concerns about human-machine collisions, particularly in nonline-of-sight (NLoS) scenarios such as corners. Developing a robot capable of locating and tracking humans behind the corners will greatly mitigate risk. However, most of them cannot work in complex environments or require a costly infrastructure. This article introduces a solution that uses the reflected and diffracted millimeter wave (mmWave) radio signals to detect and locate targets behind the corner. Central to this solution is a localization convolutional neural network (L-CNN), which takes the angle-delay heatmap of the mmWave sensor as input and infers the potential target position. Furthermore, a Kalman filter is applied after L-CNN to improve the accuracy and robustness of estimated locations. A red-green-blue-depth (RGB-D) camera is attached to the mmWave sensor as the annotation system to provide accurate position labels. The results of the experimental evaluation demonstrate that our data-driven approach can achieve remarkable positioning accuracy at the 10-cm level without extensive infrastructure. In particular, the approach effectively mitigates the adverse effects of diffraction and multibounce phenomena, making the system more resilient.
Original language | English |
---|---|
Pages (from-to) | 38102-38112 |
Number of pages | 11 |
Journal | IEEE Sensors Journal |
Volume | 24 |
Issue number | 22 |
Early online date | 23 Aug 2024 |
DOIs | |
Publication status | Published - 2024 |
MoE publication type | A1 Journal article-refereed |
Keywords
- Cameras
- Indoor positioning
- Millimeter wave communication
- Optical imaging
- Optical sensors
- Radar tracking
- Robot sensing systems
- Robots
- angle-delay estimation
- convolutional neural network
- cross-modal training
- frequency-modulated continuous-wave radar
- nonline-of-sight tracking
- robotics
- Angle-delay estimation
- nonline-of-sight (NLoS) tracking
- convolutional neural network (CNN)
- indoor positioning
- frequency-modulated continuous-wave (FMCW) radar