Directionally Controlled Time-of-Flight Sensors: Algorthims, Optical and Imaging Strategies
Abstract
We explore the novel imaging capabilities induced by a depth sensor that can control exactly where and when measurement occurs. Such a sensor is freed from the constraint of spatial samples in a fixed array and can, in real-time, adaptively sample the scene to achieve vision for complex scenes. The availability of fast microelectromechanical (MEMS) mirrors and fast computation has converged in the present moment, allowing this research to happen, for the first time, unconstrained by significant hardware limitations. Our testbed uses a MEMS mirror to reflect a single pulsed laser over a field-of-view to obtain 3D scans. The voltages that control the MEMS actuators allow analog (continuous) time-of-flight (TOF) sensing angles. As a modulator, MEMS mirrors have well-known advantages of high-speed and fast response to control. The combination of continuous sampling, fast/controllable measurements and real-time closed loop computation create an imaging platform on which we will showcase novel sensing algorithms. sampling with power could allow high-quality miniature LIDARs for small robots, smartphones and nodes in the IoT. There is significant potential for applications beyond these, and the entire space is pregnant with possibilities. PI Koppal and co-PI Xie request support from ONR for three years, including funding for two Ph.D. student (one working on optics and MEMS and the other working on algorithms and optics), resources to purchase sensors and fabrication support. Two month of summer support for the PI and co-PI for each year is also requested.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jul 26, 2018
- Source ID
- N000141812663
Entities
People
- Sanjeev J. Koppal
Organizations
- Office of Naval Research
- United States Navy
- University of Florida