Search for: All records

Abstract Purpose of ReviewWe review recent advances in algorithmic development and validation for modeling and control of soft robots leveraging the Koopman operator theory. Recent FindingsWe identify the following trends in recent research efforts in this area. (1) The design of lifting functions used in the data-driven approximation of the Koopman operator is critical for soft robots. (2) Robustness considerations are emphasized. Works are proposed to reduce the effect of uncertainty and noise during the process of modeling and control. (3) The Koopman operator has been embedded into different model-based control structures to drive the soft robots. SummaryBecause of their compliance and nonlinearities, modeling and control of soft robots face key challenges. To resolve these challenges, Koopman operator-based approaches have been proposed, in an effort to express the nonlinear system in a linear manner. The Koopman operator enables global linearization to reduce nonlinearities and/or serves as model constraints in model-based control algorithms for soft robots. Various implementations in soft robotic systems are illustrated and summarized in the review. 

There is an urgent need for low-cost and simple-to-use tools for identifying substandard and falsified medicines. In this work we demonstrate “Disintegration Fingerprinting” (DF), a technique that identifies pills, tablets, caplets, and other solid-dosage drugs based on how the drug disintegrates and dissolves in liquid. The DF hardware consists of a water-filled transparent plastic cup atop a conventional magnetic stirrer. An inexpensive sensor mounted on the outside of the cup shines infrared light into the cup and measures the amount of light that is reflected back to the sensor. When a pill is added to the stirred water, the pill begins to disintegrate into particles that swirl around inside the cup. Whenever one of these particles passes near the infrared sensor, the particle reflects additional light back to the sensor and creates a millisecond-duration peak in a plot of sensor output vs. time. The number of particles in the water changes over time as the particles continue to disintegrate and (in some cases) eventually dissolve away. By plotting the number of particles detected vs. time, we create a Disintegration Fingerprint that can be used to identify the drug product. In a proof-of-concept study, we used DF to analyze 96 pills from 32 different drug products (including antibiotics, opioid and non-opioid analgesics, antidepressants, anti-inflammatories, antiemetics, antihistamines, decongestants, muscle relaxants, expectorants, sleep aids, cold medicines, antacids, hormonal birth control, and dietary supplements, as well as a simulated falsified drug product). We found that DF correctly identified 90% of these pills, and the technique can even distinguish name-brand and generic versions of the same drug. By providing a fast (60-minute), inexpensive ($33 USD), and easy-to-use tool for identifying substandard and falsified medicines, Disintegration Fingerprinting can play an important role in the fight against fake drugs. 

Facial recognition technology is becoming increasingly ubiquitous nowadays. Facial recognition systems rely upon large amounts of facial image data. This raises serious privacy concerns since storing this facial data securely is challenging given the constant risk of data breaches or hacking. This paper proposes a privacy-preserving face recognition and verification system that works without compromising the user’s privacy. It utilizes sensor measurements captured by a lensless camera - FlatCam. These sensor measurements are visually unintelligible, preserving the user’s privacy. Our solution works without the knowledge of the camera sensor’s Point Spread Function and does not require image reconstruction at any stage. In order to perform face recognition without information on face images, we propose a Discrete Cosine Transform (DCT) domain sensor measurement learning scheme that can recognize faces without revealing face images. We compute a frequency domain representation by computing the DCT of the sensor measurement at multiple resolutions and then splitting the result into multiple subbands. The network trained using this DCT representation results in huge accuracy gains compared to the accuracy obtained after directly training with sensor measurement. In addition, we further enhance the security of the system by introducing pseudo-random noise at random DCT coefficient locations as a secret key in the proposed DCT representation. It is virtually impossible to recover the face images from the DCT representation without the knowledge of the camera parameters and the noise locations. We evaluated the proposed system on a real lensless camera dataset - the FlatCam Face dataset. Experimental results demonstrate the system is highly secure and can achieve a recognition accuracy of 93.97% while maintaining strong user privacy. 

2024 33rd IEEE International Conference on Robot and Human Interactive Communication (RO-MAN) - Late Breaking Reports 

Pneumatic control systems are common in manufacturing, healthcare, transportation, robotics, and many other fields. Undetected failures in pneumatic systems can have serious consequences. In this work, we present an air-powered error detector that can identify failures in pneumatic systems. This device contains a pneumatic logic circuit of 21 microfluidic valves that calculates the parity bit corresponding to several pneumatic control bits. If a problem such as an air leak or blockage occurs, then the calculated and expected parity bits will not match, and the device outputs an error signal to alert the user or to shut down the system. As a proof of concept, we used the device to detect anomalies in an intermittent pneumatic compression (IPC) medical device. By providing a simple and low-cost way to detect problems without using sensors, the pneumatic error detector can promote safety and reliability across a wide range of pneumatic systems.