"Every complex problem has a solution which is simple, direct, plausible - and wrong." So said US journalist H.L. Mencken. We often find ourselves negotiating this challenge when seeking to exploit weather information to improve safety and efficacy. It applies no less in the world of #autonomous driving.
A friend recently insisted, "...but, really, why would an autonomous vehicle need to know anything at all about the weather?!". Faced with a pretty simple question, there isn't always a pithy answer. I often think of it as two sets of challenges:
In taking a human driver out of the loop, you lose at least two fundamental roles: 1) perception of the vehicle's surroundings, and 2) judgment required to make decisions about the driving task. Both of these are essential to safety and both are affected by the weather, often in complex and interdependent ways.
Met Office and National Physical Laboratory (NPL) have released a new discussion paper on some of the practical challenges of meteorology for sensor systems. This builds on our Centre for Connected and Autonomous Vehicles-funded Sensor Assurance Framework collaboration which seeks to better understand weather impacts on sensors, and how that understanding can inform virtual and physical testing to ensure #CAV safety. Credit to Dave Jones for his work on this paper and for continuing to blaze the trail in exposing and distilling the complexities of weather and climate impacts in support of #FutureMobility.
The Automated Vehicles (AV) Act came into law in the UK on Monday; #selfdriving vehicles could be on UK roads by 2026. At the Met Office we are committed to collaborating across all sectors to support the safe introduction of autonomous systems of all modes (land, air, sea, subsea, space), and to imbue their development with meteorological expertise and data that improves safety and efficiency.
Met Office blog: https://lnkd.in/ePE4imuq
Discussion Paper: https://lnkd.in/eTAJG8GN
How do autonomous vehicles (AV) react to the weather?
We are working together with National Physical Laboratory (NPL) on an ongoing study, funded by the UK’s Centre for Connected and Autonomous Vehicles (CCAV), to establish a reliable framework for understanding how well AV perception sensors – the eyes and ears of the vehicle - perform in different weather conditions. Once developed, this Sensor Assurance Framework (SAF) will support validation, safety assurance and simulation testing of AVs across the UK.
The impact of weather on sensor performance is complex and we have just released a new discussion paper exploring the practical challenges of understanding the performance in different conditions. This discussion paper comes as the Automated Vehicles Act became law this week and paved the way for self-driving vehicles to possibly be on the UK's roads as soon as 2026.
Met Office Observations Principal Consultant Dave Jones says: “This discussion paper gives examples of how this complexity reveals itself on our testbed and how we might begin to handle this by careful consideration of uncertainty. We are very interested in hearing views from everyone in the wider AV community involved in safety assurance.”
Find out more in our blog: https://bit.ly/4arln4t
Read the discussion paper in full: https://bit.ly/3Kqmze5
-
Immediately looking out for sales or business development role | Recent experience in MENA market |
3moHello, I hope this message finds you well. Have you heard about the upcoming Qatar AI Summit that will take place on May 22nd in Doha? I am the head of Sales at Zeal Forum, an AI event company based in Dubai. We are organizing a series of AI conferences across the MENA region. Our upcoming event in Qatar is aimed at creating a benchmark of best practices for attendees, sponsors, and partners. Is your organization interested in exploring potential markets in Qatar, UAE, Saudi Arabia, and Bahrain? If yes, we have events lined up in these regions throughout the year. Would you be available for a quick call to discuss this opportunity? Please let me know your thoughts. Best regards, Shreyans