Schematics of the PCB module In my previous post, Developing a Fall Detection Device with Raspberry Pi, I delved into the initial steps of building a fall detection system using the Raspberry Pi Zero 2W. Today, I’m excited to share the next phase of this journey: designing and fabricating a custom PCB that integrates essential sensors to create a compact, robust, and efficient fall detection device. The Quest for Compactness The primary goal of this project was to shrink the device’s size, making it thinner and more wearable for our target users. Bulky prototypes are a hindrance, especially when designing for wearables that need to be unobtrusive yet reliable. By creating a custom PCB tailored to the Raspberry Pi Zero 2W, I aimed to eliminate the tangle of wires and breadboards, resulting in a sleek, sturdy unit. Sensor Selection: More Than Just Data Points Choosing the right sensors was crucial. The device needed to accurately detect falls by monitoring various physiological and environmental parameters. Here’s what made the cut: Accelerometer: Captures sudden changes in acceleration, a key indicator of falls. Gyroscope: Detects deviations in orientation and angular velocity. Barometer: Measures pressure changes to calculate shifts in altitude. ECG (Electrocardiogram): Monitors...
Your browser does not support the audio element. ** Dive into an AI-generated podcast where two virtual hosts discuss the key findings and implications of the featured article and its groundbreaking research." The prototype with the cover removed Falls are a significant concern for the elderly population, often leading to serious injuries and a decrease in the quality of life. Detecting falls promptly can enable quick assistance, potentially reducing the severity of injuries and providing peace of mind for both seniors and their families. In my recent project, I set out to create a highly accurate, real-time fall detection device that minimizes false positives while operating on a resource-constrained platform. Project Overview The primary goal of this project was to develop a compact and efficient fall detection system using a Raspberry Pi Zero 2W, equipped with MPU-9250 (accelerometer and gyroscope) and BMP-388 (barometric pressure) sensors. The device is designed to collect data on falls and Activities of Daily Living (ADL), process the data using Transformer-based deep learning models, and detect falls with high accuracy. A key feature of the device is its placement on the chest, near the heart. This strategic location offers several advantages: Enhanced Motion Detection: The chest...
In my last article I wrote about how I wrote Telegram Chat for tracking co-working spends. I wrote a chatbot using Java, hosted it on my Raspberry PI 3. Yes, it’s serving its needs, we fully rely on it. However, since this chatbot was written as an experiment, I wanted to proceed with research and investigate how real chatbots are written. Why Raspberry PI 3 is not the the best place for hosting a Chat Bot Since Telegram Chat Bot can use Long-Polling model, we don’t care about a dedicated IP address or DNS record, the only thing we need is the internet connection. So yes, Raspberry PI device is a computer, we can run our program there, so hosting a Telegram Chat with Long-Polling model is an option. However, what if we are writing a Chat Bot with hundreds or thousands of requests/second? Obviously, Long-Polling model could become a bottleneck since it allows to run a single application in a single Raspberry PI which means we can’t spread the load across multiple devices, which means you can’t scale the whole construction horizontally, only vertically, by adding more CPU/RAM power. Long story short, if we want to have a scalable...
I guess, everyone know knows what are chat bots. Do do I, I was looking for some idea of chat bots for many months and finally, I ended up with an idea for my co-working, which I made with my friends and where I am currently spending most of my life. During the month, we collect a list of expenses, then sum them up, divide by the number of people working in the office and then, finally, reach to every individual and ask to pay for general expenses. We were tracking expenses in the Google Spreadsheet and it was working fine, in general. However, I noticed that in most of the cases, people are too lazy to go into the spreadsheet and put their expenses there. So my obsession of being up to date with modern technologies and a tiny small problem with office expenses met together, I invested 3-4 hours of my time on writing a chat bot and now we are using it. I decided to stick with Telegram, a messenger which became very popular within the last years. I used Java as a programming language. Telegram has an API for chat bots, so I made a research...
How it started Besides my main work on Upwork I quite often pick different projets. So I found a project, where I had to write a program for recognizing chicken eggs on a factory stream line. Customer wanted to install the application on computer with web camera, put this camera at a top of stream line and the application had to calculate eggs and send them to the DB. He also wanted to run this program on a cheap computer. The quality of the network in the factory isn’t stable, so the program had to be resilient to outstand the network issues. There was enough challenges for me, so I decided to participate on this project. The biggest challenge here was that I had no serious experience with OpenCV and image recognition, so I wanted to test myself if I can deep dive into unknown field and return with successful result. Customer wanted to have 99% of recognition. This whole post will be a story how this application was designed, how it was written and what problems did I faced during the development. I will try to explain each architecture decision, from the beginning and to the end of the...
