Weekly Robotics #83

Issue 83

Two of the WR readers had mentioned the Open Source Ventilator Project - a device that might come useful for COVID-19 patients that are needing intensive care, especially if we run out of medically certified devices. I think the problem is incredibly difficult, therefore if you have experience in working on medical devices I’m sure your help would be appreciated. If you would like to get involved I think this Hackaday post might be a good entry for a discussion with other DIYers.

On a related note the virus is causing some companies to close. If you find yourself looking for work because of the virus, write me an e-mail. For the next couple of weeks I’ll be hosting a Job Seekers section - a place for you to share your profile and hopefully help you find the next job. The most clicked last week was the Open TOF LiDAR with 12.4% opens.

1) This Drone Can Play Dodgeball – And Win.
UZH
INFO: The Researchers from the Robotics and Perception Group at the University of Zurich keep improving their dynamic obstacle avoidance using event cameras. This work looks like a follow up from the work I featured in the issue #39 titled Event Camera Helps Drone Dodge Thrown Objects.

2) Drone Obstacle Avoidance in 2003: 264 Pixels and 8-Bit Processors.
DIY Drones
INFO: Since we are on the topic of obstacle avoidance: here is an interesting write up on implementing obstacle avoidance in 2003 by Geoffrey L. Barrows.

3) LIBRE-dataset.
Nagoya University
INFO: Researchers at Nagoya University had created a dataset comparing 12 multi-channel LiDARS from Velodyne, Hesai, Ouster, RoboSense and LeiShen across various scenarios (static, moving vehicle) and conditions (fog, rain, strong light). The dataset also analyses the interference when running multiple LiDARS at the same time. For more information about this dataset you can see this publication.

4) Robotics Expert Breaks Down Robot Scenes From Film & TV.
YouTube (Wired)
INFO: In this video Chris Atkeson, a professor at Carnegie Mellon University, breaks down clips from movies and TV about AI and Robotics. Although you might know most of these things already I find this video to be a good watch, I especially liked the Chernobyl robot and radiation impact that you can see towards the end of the video.

5) rex-gym.
GitHub
INFO: This repository contains OpenAI Gym environments for SpotMicro, an open source quadruped robot that was first published on ThingiVerse and then continued development was picked up by other developers and is hosted on this website.

6) Autonomous Solar Powered Rover.
YouTube (rctestflight)
INFO: This is the episode 1 of the vlog on building the solar powered rover. The rover autopilot stack used in this project is ArduRover.

7) Publication of the Week - The Marathon 2: A Navigation System (2020).
arXiv
INFO: This paper presents Navigation2, a ROS2 library for, you’ve guessed it, navigation! Navigation2 is built upon BehaviorTree.CPP making it very flexible. Since behavior trees are defined in xml you can easily change the behaviour at run time. I’m really excited to see how this this software stack will evolve, especially that I’ll be using it a lot

1) SwitchBlox - The Open Source 5-Port Ethernet Switch for Robots and Drones.
BotBlox
INFO: SwitchBlox is a small, open source Ethernet switch with five 100/100M ports designed for size constrained and tough environments. With a size of 44.5mm (1.75 inches) by 44.5mm (1.75 inches), SwitchBlox is the smallest network switch in the world, making it perfect for robots, drones or any space limited network application.

2) Humble Book Bundle: Coding Starter Kit by No Starch Press.
Humble Bundle
INFO: Ready, set, code! Designed for kids, this bundle of ebooks is the perfect first step for anyone ready to jump into the world of coding! As usual with Humble Bundles featured in this newsletter you can choose to support Weekly Robotics at checkout.

Job Seekers

Name: Shantanu Wadnerkar
Location: Davis, CA, USA. Willing to relocate within USA
Skills: Python, Linux, MATLAB/Simulink, C++, ROS, numpy, matplotlib, openai-gym
Profile: I’m a Robotics Engineer working with a myriad of sensors both in simulation and hands-on. Currently working on Baxter research robotic arm with Intel 3D camera. Also, contributing to an open-source self-driving vehicle simulator
Social Profiles: LinkedIn
Email: smwadnerkar5@gmail.com


Name: Gasser Abdelnasser Ahmed
Location: Cairo, Egypt. Willing to Relocate
Skills: SolidWorks, Matlab, Simulink, Python, C, ROS ,Control, Robotics, Autonomous Systems
Profile: I am a Robotics Engineer who is obsessed with Autonomous Systems, Robotics and Control,Looking for a company which is interested in developing Robotics solutions or Autonomous Systems
Social Profiles: LinkedIn
Email: gasser_nasser@yahoo.com