Development of an Automated Delivery Vehicle

The Covid – 19 pandemic has gone out of control all over the world causing nearly five million deaths. Number of affected persons keeps increasing and there is a huge problem of providing essential items for them. The objective of this project is to design an Automated Delivery System, which is suitable for Sri Lanka during such pandemic situations. Because of the situations like this, delivery people do not like to reach affected areas even for business. Therefore, the people living in such areas face the problem of getting even the basic needs such as essential food items and medicine. As a solution, an unmanned automated delivery vehicle system has been proposed in this research, which can be mainly used in locked down areas in Sri Lanka. It can be used to deliver goods without a driver to the required places, while maintaining the health guidelines and required security and safety. This can be further extended to certain areas such as private apartments, hospitals, supermarkets etc. to deliver items in automated way.

The Algorithm of a Game-Based System in the Relation between an Operator and a Technical Object in Management of E-Commerce Logistics Processes with the Use of Machine Learning

Machine learning (ML) is applied in various logistic processes utilizing innovative techniques (e.g., the use of drones for automated delivery in e-commerce). Early challenges showed the insufficient drones’ steering capacity and cognitive gap related to the lack of theoretical foundation for controlling algorithms. The aim of this paper is to present a game-based algorithm of controlling behaviours in the relation between an operator (OP) and a technical object (TO), based on the assumption that the game is logistics-oriented and the algorithm is to support ML applied in e-commerce optimization management. Algebraic methods, including matrices, Lagrange functions, systems of differential equations, and set-theoretic notation, have been used as the main tools. The outcome is a model of a game-based optimization process in a two-element logistics system and an algorithm applied to find optimal steering strategies. The algorithm has been initially verified with the use of simulation based on a Bayesian network (BN) and a structured set of possible strategies (OP/TO) calculated with the use of QGeNie Modeller, finally prepared for Python. It has been proved the algorithm at this stage has no deadlocks and unforeseen loops and is ready to be challenged with the original big set of learning data from a drone-operating company (as the next stage of the planned research).

670 Changes in Childhood Sleep Patterns in an Intervention Study Prior to and During COVID19 Restrictions

Introduction We conducted a childhood sleep promotion study between March 2019 and December 2020 in Philadelphia. COVID19 was first detected in Pennsylvania in March 2020 and non-essential services were strictly curtailed (including school closures), with easing of curtailments by the fall 2020 (including hybrid schooling in some districts). We determined if changes in sleep duration were consistent during pre-, earlier, and later COVID19 periods. Methods Typically developing children (9-12y) with sleep duration <8.5 hours per weeknight were enrolled. Sleep was measured using Fitbit devices during a baseline week and a 7-week intervention period. A factorial design was used to test five candidate intervention components: 1) sleep goal; 2) electronic device reduction messaging; 3) daily routine messaging; 4) child-directed financial incentive; and 5) parent-directed financial incentive. Sleep data were transmitted to a mobile health platform that automated delivery of the intervention components. We categorized participants when they completed the study: 1) Spring-Fall 2019 semesters (pre-COVID19); 2) Spring 2020 semester (started pre-COVID19, with strict restrictions impacting intervention periods); or 3) Fall 2020 semester (easing of COVID19 restrictions). Mixed effect modelling determined sleep changes. Results Mean age of participants was 11.6y (51% female and 29% Black participants). Pre-COVID19 (N=59), average sleep duration increased from baseline by 21 (95% CI: 10, 30) minutes per weeknight during the intervention. In spring 2020 (N=18), the average sleep duration increase was two times larger in magnitude at 41 (95% CI: 25, 59) minutes per weeknight. For fall 2020 (N=20), the average sleep duration increase was 24 (95% CI: 7, 40) minutes per weeknight. Changes in sleep timing from baseline during the intervention were consistent pre-COVID19 and in the fall 2020 (e.g., ≈15 minutes earlier sleep onset throughout the intervention period), whereas sleep timing changes were dynamic in the spring 2020 (e.g., 41 minutes earlier for week 1, and 44 minutes later for week 7). Conclusion This sleep intervention demonstrated increases in sleep duration pre-COVID19, with marked duration increases and dynamic timing changes coinciding with COVID19 restrictions during earlier (Spring 2020), but not later (Fall 2020), weeks of the COVID19 pandemic in Pennsylvania. Support (if any) K0 1 HL1 2 3 6 1 2 and CHOP

A “Smart” Biosensor-Enabled Intravascular Catheter and Platform for Dynamic Delivery of Propofol to “Close the Loop” for Total Intravenous Anesthesia

ABSTRACT Background Target-controlled infusion anesthesia is used worldwide to provide user-defined, stable, blood concentrations of propofol for sedation and anesthesia. The drug infusion is controlled by a microprocessor that uses population-based pharmacokinetic data and patient biometrics to estimate the required infusion rate to replace losses from the blood compartment due to drug distribution and metabolism. The objective of the research was to develop and validate a method to detect and quantify propofol levels in the blood, to improve the safety of propofol use, and to demonstrate a pathway for regulatory approval for its use in the USA. Methods We conceptualized and prototyped a novel “smart” biosensor-enabled intravenous catheter capable of quantifying propofol at physiologic levels in the blood, in real time. The clinical embodiment of the platform is comprised of a “smart” biosensor-enabled catheter prototype, a signal generation/detection readout display, and a driving electronics software. The biosensor was validated in vitro using a variety of electrochemical methods in both static and flow systems with biofluids, including blood. Results We present data demonstrating the experimental detection and quantification of propofol at sub-micromolar concentrations using this biosensor and method. Detection of the drug is rapid and stable with negligible biofouling due to the sensor coating. It shows a linear correlation with mass spectroscopy methods. An intuitive graphical user interface was developed to: (1) detect and quantify the propofol sensor signal, (2) determine the difference between targeted and actual propofol concentration, (3) communicate the variance in real time, and (4) use the output of the controller to drive drug delivery from an in-line syringe pump. The automated delivery and maintenance of propofol levels was demonstrated in a modeled benchtop “patient” applying the known pharmacokinetics of the drug using published algorithms. Conclusions We present a proof-of-concept and in vitro validation of accurate electrochemical quantification of propofol directly from the blood and the design and prototyping of a “smart,” indwelling, biosensor-enabled catheter and demonstrate feedback hardware and software architecture permitting accurate measurement of propofol in blood in real time. The controller platform is shown to permit autonomous, “closed-loop” delivery of the drug and maintenance of user-defined propofol levels in a dynamic flow model.

Alpha-N-V2: Shortest Path Finder Automated Delivery Robot with Obstacle Detection and Avoiding System

An improved version of Alpha-N, a self-powered, wheel-driven Automated Delivery Robot (ADR), is presented in this study. Alpha-N-V2 is capable of navigating autonomously by detecting and avoiding objects or obstacles in its path. For autonomous navigation and path planning, Alpha-N uses a vector map and calculates the shortest path by Grid Count Method (GCM) of Dijkstra’s Algorithm. The RFID Reading System (RRS) is assembled in Alpha-N to read Landmark determination with Radio Frequency Identification (RFID) tags. With the help of the RFID tags, Alpha-N verifies the path for identification between source and destination and calibrates the current position. Along with the RRS, GCM, to detect and avoid obstacles, an Object Detection Module (ODM) is constructed by Faster R-CNN with VGGNet-16 architecture that builds and supports the Path Planning System (PPS). In the testing phase, the following results are acquired from the Alpha-N: ODM exhibits an accuracy of [Formula: see text], RRS shows [Formula: see text] accuracy and the PPS maintains the accuracy of [Formula: see text]. This proposed version of Alpha-N shows significant improvement in terms of performance and usability compared with the previous version of Alpha-N.