Motional’s Post

Hey everyone! Here's a short update on the traffic system that I have been working on. Last week we had setup the graph system with nodes and edges. I have added cars (blue circles for now) to the system. The cars spawn at one of the nodes, decide on a random destination node, find the shortest path to the destination, and proceed to move towards each node in the path. You can try out the demo here https://lnkd.in/gRD8SvTu GitHub : https://lnkd.in/gCPXY6ZJ A quick thought on the next steps would be to implement a type of queueing system for the cars, so that they don't race against one another and wait for their turn. One way of doing this would be to implement node occupancy. So a node can be occupied by only one car at a time and when the car leaves the node, it becomes free to be occupied by another car. Each car would wait for the next node in the path to become free, before moving towards it. This brings up an interesting perspective. In the last update we thought of nodes as intersections and edges as roads, but now I have a different view to this, which I'll explain more in my next update. Do let me know your guesses on what this new perspective would be. Cheers! #buildinpublic #gamedev #trafficsimulation

As technological innovations speed onto our highways 🛣️, we're witnessing a major shift in driving safety. Our latest blog delves into the advanced systems and devices, powered by optical components, that are reshaping our roads. Discover how AI is taking these innovations to the next level, making our journeys smarter and safer. 🚘 Read our latest blog, 'Road Safety Revolution: How AI and Optical Components Are Joining Forces to Tackle Driving Offences in the UK', here: https://bit.ly/4bt2Qqr

I'm moderating this webinar Thursday. Free for employees of TRB sponsors and Titanium and Cobalt Global Affiliates (sorry, I don't know who those are); $100 for others. #RoadSafety #VisionZero #cycling #walking #ZeroTrafficDeaths #driving #DrivingSafety #SaferStreets #SaferStreetsForAll #simulation #research #transportation #lighting #conspicuity #TranspoTech

How does Motion Detection 2.0 work? Hikvision Motion Detection 2.0 features person & vehicle target classification, easy configuration, and efficient playback. Person and vehicle target classification: In Hikvision's laboratories, deep-learning algorithms are trained to classify objects in videos into three categories – Persons, Vehicles, and Others. Users can further configure the algorithms to automatically detect persons, vehicles, or both based on their specific requirements and then notify them in real-time. Learn More: https://lnkd.in/gPQCZSwM

Reflections on Using Mirrors When Driving. https://hubs.ly/Q02d-Lbd0

Vehicle Counting in Parking Regions Using Ultralytics YOLOv8 🔥🔥 🔗 Code Link: https://lnkd.in/gAYYkVku 🔗 Real-world Examples: https://lnkd.in/g2j8CiK7 ✅ Counting in the region can be used in many real-world problems 🚀  ✅ Model is not trained on the large dataset, but still performing well🔥 🚀 Offer insights in the parking area, more features are coming soon 💪 For more information, you can check our Docs: https://lnkd.in/dRj5XyPx #computervision #objectdetection #parkingmanagement #yolov8 Glenn Jocher Paula Derrenger

My recent publication is now available online. This effort emphasizes the differences in driver reaction time across the road network. The reaction time on curves exceeds the standard reaction time by 4% - 12%. This is not the whole story yet it is another step in my journey of understanding driver behavior and how the driver interacts with the surrounding environment. https://lnkd.in/g9Rcmu2j

If you follow autonomy or autonomous cars/robots, you know that they need a precision location to navigate safely. I think Point One Navigation is being a little shy here about how ground breaking this piece of technology is... affordable, reliable precision location will unlock all sorts of applications. Competitive products are 10x the cost (at least) and won't enable scale. The Atlas will.