We present PRM-RL, a hierarchical method for long-range navigation task completion that combines samplingbased path planning with reinforcement learning (RL). The RL agents learn short-range, point-to-point navigation policies that capture robot dynamics and task constraints without knowledge of the large-scale topology. Next, the sampling-based planners provide roadmaps which connect robot configurations that can be successfully navigated by the RL agent. The same RL agents are used to control the robot under the direction of the planning, enabling long-range navigation. We use the Probabilistic Roadmaps (PRMs) for the sampling-based planner. The RL agents are constructed using feature-based and deep neural net policies in continuous state and action spaces. We evaluate PRM-RL, both in simulation and on-robot, on two navigation tasks with non-trivial robot dynamics: end-to-end differential drive indoor navigation in office environments, and aerial cargo delivery in urban environments with load displacement constraints. Our results show improvement in task completion over both RL agents on their own and traditional sampling-based planners. In the indoor navigation task, PRM-RL successfully completes up to 215 m long trajectories under noisy sensor conditions, and the aerial cargo delivery completes flights over 1000 m without violating the task constraints in an environment 63 million times larger than used in training.
We collected the respiratory mucus coating the endotracheal tubes used during short surgical procedures in 27 patients with no clinical evidence of respiratory disease. Twelve were male and 15 were female, and they ranged from 1 to 64 yr of age (mean, 28.7 yr). The viscoelastic properties, frog palate transport rate, and percent solid composition were in the normal range reported for both canine and human mucus collected using the bronchoscopy brush technique. There were no significant differences noted between male and female patients, and there were no changes in mucus or transport properties seen with aging. Mucus was also collected separately from the inside of the tube exposed to constant gas flow (13 patients), and from the outside of the endotracheal tubes in 25 patients. Although there were no significant differences in viscoelastic properties between inside and outside mucus, there was a greater thread formation (filance, 45 versus 26 mm; p less than 0.005) and a higher percentage of solids in mucus from the inside of the endotracheal tube (15.9 versus 11.4%; p less than 0.05), which is compatible with reduced hydration. The duration of anesthesia ranged from 25 to 195 min (mean, 85 min). There was no effect of duration of anesthesia on any of the measured mucus properties. This technique for mucus collection allows us to study alterations in mucus properties in patients with and without pulmonary disease at the time of incidental surgery.(ABSTRACT TRUNCATED AT 250 WORDS)
To better understand the frog palate model of mucociliary transport, we measured the transport rate of mucus (MTR) from the leopard frog, Rana pipiens, and from the bullfrog, R. catesbeiana, recorded the stability of the MTR over a period of hours and days and over the course of 1 yr, and measured the viscoelasticity, percent solid composition, and spinnability (filance) of mucus from both species. Bullfrog mucus was less rigid than leopard frog mucus (log G* at 1 rad/s 2.09 vs. 2.61; P less than 0.01) and had a higher viscosity-to-elasticity ratio (tan delta at 1 rad/s 0.36 vs. 0.26; P less than 0.05). It also had a lower solids content (8.71 vs. 13.72%; P = 0.02), and there was a trend to lower spinnability for bullfrog mucus (filance 26.7 vs. 33.5 mm). These data suggest that bullfrog mucus has viscoelastic properties similar to normal mammalian respiratory mucus and leopard frog mucus has viscoelasticity similar to sputum samples. MTR was significantly slower in the winter than in the summer months (17 vs. 30 mm/min; P less than 0.0001). Although the leopard frog palate could be used for at least 7 consecutive days without exhaustion, bullfrog palates could be used for only 5 days. Palates of either species could generally be tested for 6 h/day without a significant decrease in MTR. These data clarify some of the sources of variability in the use of this system and suggest methods of standardization.
Erythromycin inhibits mucus glycoconjugate secretion from airway cells in vitro and may increase mucus clearance in patients with asthma or diffuse panbronchiolitis. Because mucus hypersecretion is common in purulent rhinitis, we questioned whether clarithromycin would change the properties of nasal mucus from subjects without sinus disease and from patients with acute purulent rhinitis. Nasal secretions were collected before and after nasal methacholine challenge from 10 adults without nasal symptoms and without methacholine from 10 patients with purulent rhinitis. After 2 wk of oral clarithromycin (500 mg twice daily), secretions were again collected from both groups. Secretions were analyzed for viscoelasticity, cohesion, hydration, and ciliary and airflow (sneeze) transportability. Compared with secretions from healthy subjects, rhinitis secretions had decreased wettability (contact angle on Teflon 100 degrees versus 84.67 degrees; p = 0.001), increased cohesion (36.8 versus 24.3 mm; p = 0.003), decreased sneeze clearability (20.6 versus 32 mm; p = 0.04), and increased percent solids (4.61 versus 2.82%; p = 0.04). After clarithromycin, the rheology, hydration, cohesion, and transportability of the rhinitis secretions were similar to those of the postclarithromycin secretions from the healthy subjects. Secretion volume also decreased (500.1 versus 28.3 mg; p = 0.01), and mucociliary transportability increased by 30% (0.76 versus 0.99; p = 0.005). Although clarithromycin reduced mucus secretion in both rhinitis patients and normal subjects, it did not alter the secretory response to methacholine.
Neck injuries and the related pain have a high prevalence and represent an important health problem. To properly diagnose and treat them, practitioners need an accurate system for measuring Cervical Range Of Motion (CROM). This article describes the development and validation of an inexpensive, small (4 cm × 4 cm × 8 cm), light (< 200 g) and easy to use solution for measuring CROM using wearable inertial sensors. The proposed solution has been designed with the clinical practice in mind, after consulting with practitioners. It is composed of: (a) two wearable wireless MEMS-based inertial devices, (b) a recording and report generation software application and (c) a measurement protocol for assessing CROM. The solution provides accurate (none of our results is outside the ROM ranges when compared with previously published results based on an optical tracking device) and reliable measurements (ICC = 0.93 for interrater reliability when compared with an optical tracking device and ICC > 0.90 for test-retest reliability), surpassing the popular CROM instrument’s capabilities and precision. It also fulfills the needs for clinical practice attending to effectiveness, efficiency (4 min from setup to final report) and user’s satisfaction (as reported by practitioners). The solution has been certified for mass-production and use in medical environments.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.