-
Novel Phase Detector Measurement Procedure Using Quasi-Synchronized RF Generator
Authors:
V. A. Pulido,
F. Cabrera-Almeida,
P. Quintana-Morales,
E. Mendieta-Otero
Abstract:
This paper presents a new procedure for phase detector measurements that allows the use of generators that share a 10 MHz reference oscillator but do not synchronize in phase, in other words, quasi-synchronized RF generators. The objectives are taking advantage of the benefits of using two generators but recovering lower-cost generators that have worse synchronization performance and opening the d…
▽ More
This paper presents a new procedure for phase detector measurements that allows the use of generators that share a 10 MHz reference oscillator but do not synchronize in phase, in other words, quasi-synchronized RF generators. The objectives are taking advantage of the benefits of using two generators but recovering lower-cost generators that have worse synchronization performance and opening the door to the possibility of using a very simple control element based in Arduino Uno and cheaper instruments. The new procedure is characterized by continuously alternating calibration and measurement sequences to make up for the phase drift of quasisynchronized generators and guarantee a maximum phase error specification (+-1 grade in this paper). Data acquisition has been divided in two stages: measurement of detector curves without phase reference (in-phase and phase-shifted) and measurement of reference data. All the data is later combined to obtain correctly referenced in-phase detector curves. The technique can be reproduced with other equivalent instrumentation. The novel procedure that allows compensation for errors (amplitude, phase shift, mismatching, etc.) is detailed, and its relation to the required measurement accuracy is amply discussed. The proposed technique is applied to characterize a phase detector based on in-phase and phase-shifted multiplication from 3 to 8 GHz with 1 GHz step. Measurements have a final maximum error of +-2 grade for both frequency and calibrated input power, according to the accuracy specifications of the VNA used to calibrate the signal distribution network, added to the +-1 grade specified in this new procedure.
△ Less
Submitted 7 February, 2024;
originally announced February 2024.
-
Uncertainty-Aware Guidance for Target Tracking subject to Intermittent Measurements using Motion Model Learning
Authors:
Andres Pulido,
Kyle Volle,
Kristy Waters,
Zachary I. Bell,
Prashant Ganesh,
Jane Shin
Abstract:
This paper presents a novel guidance law for target tracking applications where the target motion model is unknown and sensor measurements are intermittent due to unknown environmental conditions and low measurement update rate. In this work, the target motion model is represented by a transformer neural network and trained by previous target position measurements. This transformer motion model se…
▽ More
This paper presents a novel guidance law for target tracking applications where the target motion model is unknown and sensor measurements are intermittent due to unknown environmental conditions and low measurement update rate. In this work, the target motion model is represented by a transformer neural network and trained by previous target position measurements. This transformer motion model serves as the prediction step in a particle filter for target state estimation and uncertainty quantification. The particle filter estimation uncertainty is utilized in the information-driven guidance law to compute a path for the mobile agent to travel to a position with maximum expected entropy reduction (EER). The computation of EER is performed in real-time by approximating the information gain from the predicted particle distributions relative to the current distribution. Simulation and hardware experiments are performed with a quadcopter agent and TurtleBot target to demonstrate that the presented guidance law outperforms two other baseline guidance methods.
△ Less
Submitted 20 March, 2025; v1 submitted 1 February, 2024;
originally announced February 2024.
-
Time and Cost-Efficient Bathymetric Mapping System using Sparse Point Cloud Generation and Automatic Object Detection
Authors:
Andres Pulido,
Ruoyao Qin,
Antonio Diaz,
Andrew Ortega,
Peter Ifju,
Jaejeong Shin
Abstract:
Generating 3D point cloud (PC) data from noisy sonar measurements is a problem that has potential applications for bathymetry mapping, artificial object inspection, mapping of aquatic plants and fauna as well as underwater navigation and localization of vehicles such as submarines. Side-scan sonar sensors are available in inexpensive cost ranges, especially in fish-finders, where the transducers a…
▽ More
Generating 3D point cloud (PC) data from noisy sonar measurements is a problem that has potential applications for bathymetry mapping, artificial object inspection, mapping of aquatic plants and fauna as well as underwater navigation and localization of vehicles such as submarines. Side-scan sonar sensors are available in inexpensive cost ranges, especially in fish-finders, where the transducers are usually mounted to the bottom of a boat and can approach shallower depths than the ones attached to an Uncrewed Underwater Vehicle (UUV) can. However, extracting 3D information from side-scan sonar imagery is a difficult task because of its low signal-to-noise ratio and missing angle and depth information in the imagery. Since most algorithms that generate a 3D point cloud from side-scan sonar imagery use Shape from Shading (SFS) techniques, extracting 3D information is especially difficult when the seafloor is smooth, is slowly changing in depth, or does not have identifiable objects that make acoustic shadows. This paper introduces an efficient algorithm that generates a sparse 3D point cloud from side-scan sonar images. This computation is done in a computationally efficient manner by leveraging the geometry of the first sonar return combined with known positions provided by GPS and down-scan sonar depth measurement at each data point. Additionally, this paper implements another algorithm that uses a Convolutional Neural Network (CNN) using transfer learning to perform object detection on side-scan sonar images collected in real life and generated with a simulation. The algorithm was tested on both real and synthetic images to show reasonably accurate anomaly detection and classification.
△ Less
Submitted 18 October, 2022;
originally announced October 2022.
