Abraham Bachrach

This is a perfect example of a lack of an executive command level sanity check to ensure the system is limited to viable decision options by a highly robust set of “mission level” parameters. It should KNOW this was a one-way from a map database on board and should have thrown out the left turn as even being an option. We must do better or “autonomy” is DOA. Systems need to enforce safe outcome by purposeful architecture design and hierarchy of command structures.

29

19 Comments

Interesting Autonomous car behavior. You'd think that GNSS would at least inform the car of what lane they are in and direction of traffic, while the camera/lidar would determine that all the other cars are pointed the other direction...

6

1 Comment

How it feels when everything goes wrong: (The hardships of hardware development) [𝗽𝗮𝗿𝘁 5] I have to rewind from the spring of 2023 to the fall 2022. We hired an agent to help us fundraise in the fall of 2022. The agreement was the first part of the money on our account in February. When they failed to do that, our own investors pulled the bridge financing. Now we had 3 months before we ran out of money. It was fair enough. We had struggled with motor problems for 6 months. The progress on the flying prototype seemed very little. In the end, we found out there were manufacturing errors on them. The countdown timer is 3 months now. Fugg!! Well, Travis James Mathers and I were not ready to give up. We geared up, prepared ourselves for war. I had to fire the whole team due to Danish rules around responsibility. Here is the status at this point: - A fired team with 3 months of resignation, - A prototype with motors that had been causing problems - There is no regulatory approval to fly outside. Okay, here we go! We put in all our efforts to get regulatory approval so we could fly outside and do a demonstration for a potential customer, which would probably get us our bridge financing back. 1.5 months later, we got our regulatory approval as the only ones in Europe able to do onshore and offshore operations with this heavy drone for cargo delivery for wind turbines. 2 months later, we fly outside for the second time, and at about 25 meters of height, we have a small glitch on a motor that causes instability that makes the drone crash into the ground. At this point, all the escape hatches are open for us to jump into and give up. Double FUGG!! Follow me for the next part of the story. If you love what you do, fight for it with everything you have. -MK

474

107 Comments

Its sometimes hard to believe that we live in an era where driverless vehicles aren't just a vision anymore but a reality. Waymo driverless cabs for example, are transforming the San Francisco streets forever! It was a wonderful experience working on an application of autonomous vehicles as part of my final project in the course CS267: Applications of Parallel Computing at the University of California, Berkeley. Our team, comprising Terence N., Zeid Solh, and myself, focused on parallelizing stereo vision for driverless applications. Stereo vision is crucial for self-driving cars as it builds a 3D understanding of the environment by using two cameras. However, real-time processing of these images is computationally expensive, posing a significant challenge for autonomous vehicles. We tackled this by exploring various parallel computing techniques to improve the speed of calculating the Disparity Map algorithm. We used popular techniques such as #SIMD (Single Instruction, Multiple Data), #OpenMP (multi-threading), and CUDA (NVIDIA GPU acceleration), and benchmarked the performance. By distributing the computational load across multiple cores and processors, we achieved substantial performance improvements, enabling real-time processing of high-resolution images and videos. The results were promising! Utilizing CUDA, we saw significant speedups, demonstrating the potential for efficient, real-time depth perception in autonomous systems. This project signified the importance of parallel computing in advancing autonomous driving technologies. I'm incredibly proud of our team's hard work and the insights we gained through this project. Check out our GitHub repo: https://lnkd.in/gYHrEU9z for more details on our implementation! #UCberkeley #CS267 #ParallelComputing #StereoVision #AutonomousDriving #DeepLearning #ComputerVision

49

3 Comments

This is one of the best articles written in pre-early UAM days by Peter Shannon. The productivity levels of Ancient Egyptian civilization rose to unbelievable levels as improved transportation services enhanced their economic growth leading to rise in devoting more time to progress in areas of Science, Arts and Philosophy exemplifying a progressive state of prosperity. Electric propulsion and automation — represents the same inclusion of progress into our lifestyle. https://lnkd.in/d7ad63Zg

3

2 Comments

Interesting article on Sierra Space's LIFE (essentially inflatable) modules and how they plan to make these really outsized over the course of their program. I'd love to know more about how these modules are outfitted after they are deployed. Do all the facilities need to be launched separately and integrated after the module is deployed? #commercialspace #spacestation #spaceresearch

6

Can a Parasail Turn Cargo Ships into Sailboats? ⛵ Airseas is harnessing wind energy to propel massive ships with advanced systems designed using Simulink and Model-Based Design. MathWorks CEO Jack Little explains how this engineered system works and how Airseas may transform the cargo shipping industry 👇 #ModelBasedDesign #Simulink #Engineering Learn about other customers’ technical achievements using MATLAB and Simulink: https://spr.ly/6044S5na2

1

Starpath Robotic receives $12 million funding for Lunar Water Mining   Starpath Robotics, which recently came out of stealth mode (https://lnkd.in/dZy2faaS), is focusing on a critical aspect of In-Situ Resource Utilisation (ISRU): harvesting water ice to produce liquid oxygen (LOX), a crucial component in rocket propellant. The company's innovative approach has attracted significant investment, with a recent $12 million seed round. This funding will support Starpath's rapid expansion, with plans to increase its current team of employees as well as testing of the technology in a simulated Moon environment.     Starpath's ambitious architecture involves deploying swarms of mining rovers to extract hundreds of tons of water from the lunar south-pole. The water would then be processed in autonomous lunar plants to split it into hydrogen and oxygen, and liquefy the oxygen. The entire operation would be powered by a massive solar array, developed in collaboration with New Space solar startup Solestial.   Starpath's co-founder and CEO, Saurav Shroff, envisions producing around 1,000 tons of LOX annually on the Moon. The business plan is simple: sell LOX to support regular missions by SpaceX, Blue Origin and others that are already building reusable vehicles to visit the Moon.   The startup is making swift progress, aiming to conduct a full-scale demonstration of its system in a simulated lunar environment by the end of the year. This will be followed by a series of test campaigns before launching their first demonstration mission, targeting an annual production equivalent of 100 tons of LOX.   While the challenges are significant – no entity has ever harvested or refined lunar water ice – Starpath's success could be transformative for space exploration. The company is already looking beyond the Moon, considering adaptations of its technology for Mars.   Saurav Shroff states: “We’re building commercial scale propellant production factories for the surface of the Moon and Mars. Access to abundant propellant on the surface enables fully-reusable mission profiles, where no costly spacecraft hardware is expended per-flight. This both drives down cost and enables payloads to be returned to Earth - a critical capability for servicing a high volume of crewed flights to and from the Moon and Mars.” With Starpath's innovations, the dream of sustainable space exploration may be closer to reality than ever before.   Image Credit: Starpath - Prototypes of Starpath’s lunar mining infrastructure inside the company’s offices.   #LunarMining #RocketFuel #InSituResourceUtilization #ISRU #MarsMining #SpaceLogistics

24

5 Comments

🧙‍♂️ ❗ On my quest to deploy our limited resources well, I recently (06-21-24) wrote Xometry, Fictiv, Prototype Hubs, and Jiga... The question was this: 𝙒𝙚 𝙬𝙖𝙣𝙩 𝙩𝙤 𝙗𝙚 𝙖 𝙨𝙪𝙥𝙥𝙡𝙞𝙚𝙧/𝙥𝙖𝙧𝙩𝙣𝙚𝙧 𝙛𝙤𝙧 <𝙞𝙣𝙨𝙚𝙧𝙩 𝙘𝙤𝙢𝙥𝙖𝙣𝙮>. 𝘽𝙚𝙛𝙤𝙧𝙚 𝙬𝙚 𝙥𝙪𝙧𝙘𝙝𝙖𝙨𝙚 𝙢𝙖𝙘𝙝𝙞𝙣𝙚𝙨 𝙛𝙤𝙧 𝙤𝙪𝙧 𝙨𝙝𝙤𝙥, 𝙬𝙚'𝙙 𝙡𝙞𝙠𝙚 𝙩𝙤 𝙗𝙚𝙩𝙩𝙚𝙧 𝙪𝙣𝙙𝙚𝙧𝙨𝙩𝙖𝙣𝙙 𝙬𝙝𝙖𝙩 𝙩𝙮𝙥𝙚 𝙤𝙛 𝙟𝙤𝙗𝙨 𝙮𝙤𝙪𝙧 𝙥𝙡𝙖𝙩𝙛𝙤𝙧𝙢 𝙨𝙚𝙚𝙨 𝙖𝙣𝙙 𝙞𝙛 𝙩𝙝𝙚𝙧𝙚 𝙖𝙧𝙚 𝙖𝙣𝙮 𝙖𝙧𝙚𝙖𝙨 𝙤𝙛 𝙩𝙝𝙚 𝙢𝙖𝙘𝙝𝙞𝙣𝙞𝙣𝙜 𝙢𝙖𝙧𝙠𝙚𝙩 𝙩𝙝𝙖𝙩 𝙖𝙧𝙚 𝙪𝙣𝙙𝙚𝙧𝙨𝙚𝙧𝙫𝙚𝙙 (𝙞𝙣𝙙𝙪𝙨𝙩𝙧𝙮, 𝙢𝙖𝙩𝙚𝙧𝙞𝙖𝙡, 𝙘𝙤𝙢𝙥𝙡𝙚𝙭𝙞𝙩𝙮, 𝙘𝙚𝙧𝙩𝙞𝙛𝙞𝙘𝙖𝙩𝙞𝙤𝙣𝙨, 𝙘𝙤𝙢𝙥𝙡𝙞𝙖𝙣𝙘𝙚, 𝙚𝙩𝙘.). 𝙏𝙤 𝙗𝙚 𝙘𝙡𝙚𝙖𝙧, 𝙄’𝙢 𝙝𝙤𝙥𝙞𝙣𝙜 𝙩𝙖𝙡𝙠 𝙩𝙤 𝙨𝙤𝙢𝙚𝙤𝙣𝙚 𝙩𝙤 𝙗𝙚𝙩𝙩𝙚𝙧 𝙪𝙣𝙙𝙚𝙧𝙨𝙩𝙖𝙣𝙙 𝙬𝙝𝙖𝙩 <𝙞𝙣𝙨𝙚𝙧𝙩 𝙘𝙤𝙢𝙥𝙖𝙣𝙮> 𝙣𝙚𝙚𝙙𝙨 𝙞𝙣 𝙩𝙝𝙚𝙞𝙧 𝙨𝙪𝙥𝙥𝙡𝙞𝙚𝙧𝙨. 𝙒𝙚 𝙬𝙖𝙣𝙩 𝙩𝙤 𝙙𝙚𝙨𝙞𝙜𝙣 𝙤𝙪𝙧 𝙢𝙖𝙘𝙝𝙞𝙣𝙞𝙣𝙜 𝙗𝙪𝙨𝙞𝙣𝙚𝙨𝙨 𝙛𝙧𝙤𝙢 𝙩𝙝𝙚 𝙜𝙧𝙤𝙪𝙣𝙙 𝙪𝙥 𝙩𝙤 𝙥𝙚𝙧𝙛𝙚𝙘𝙩𝙡𝙮 𝙛𝙞𝙩 <𝙞𝙣𝙨𝙚𝙧𝙩 𝙘𝙤𝙢𝙥𝙖𝙣𝙮>’𝙨 𝙚𝙭𝙥𝙚𝙘𝙩𝙖𝙩𝙞𝙤𝙣𝙨 𝙖𝙣𝙙 𝙣𝙚𝙚𝙙𝙨. To Xometry's credit, they got back to me quickly (thanks Meredith Jenkins). Meredith provided great information and made it clear that AS9100 and ITAR will be a differentiating factor. I know what the community is going to say… “You should be focusing on getting local customers in your area and building relationships.” 💯 👈 🤦‍♂️ Right, of course, we are doing that as well. But the benefit of using these platforms is that they provide some guardrails when getting started - well defined processes, software, network of customers, rules & expectations, etc. We are prepared to jump into the deep end of running a shop, but there is nothing wrong with taking a few “pool floaties” with us to make the transition a bit smoother. 🏊‍♂️ 🤷‍♂️ ⭐ 𝗜𝗳 𝘆𝗼𝘂 𝗮𝗿𝗲 𝗮 𝗽𝗲𝗿𝘀𝗼𝗻 𝘄𝗵𝗼 𝗽𝘂𝗿𝗰𝗵𝗮𝘀𝗲𝘀 𝘀𝗲𝗿𝘃𝗶𝗰𝗲𝘀 𝗳𝗿𝗼𝗺 𝗖𝗡𝗖 𝘀𝗵𝗼𝗽𝘀, 𝗺𝗮𝘆 𝗜 𝗽𝗹𝗲𝗮𝘀𝗲 𝗵𝗮𝘃𝗲 𝟭𝟱 𝗺𝗶𝗻𝘂𝘁𝗲𝘀 𝗼𝗳 𝘆𝗼𝘂𝗿 𝘁𝗶𝗺𝗲 𝘁𝗼 𝗱𝗶𝘀𝗰𝘂𝘀𝘀 𝘄𝗵𝗮𝘁’𝘀 𝗶𝗺𝗽𝗼𝗿𝘁𝗮𝗻𝘁 𝘁𝗼 𝘆𝗼𝘂 𝗶𝗻 𝗮 𝘀𝘂𝗽𝗽𝗹𝗶𝗲𝗿? My hunch is that reliability and communication are probably high on that list, but I’d love to have the conversation and talk specifics if you are willing! #CNC #Machining #Manufacturing #Reindustrialize #GCode #Startup #Quality #Supplier #Makingthings (09-03-24 edit) - While the bootstrapping idea was a noble pursuit, we've decided that our best chance to chase our "pie in the sky" dream is to leverage VC investment and partners. https://lnkd.in/e3bbwVJJ

5

1 Comment

Our Neptaverse's cloud storage expenses dropped to just $50 after we started storing the 3D models here!! ⬇ 💸 Neptaverse ( our own XR platform) previously ran completely on AWS with massive bills as s3 sucks for global customers and cloudfront is expensive. In turn , we developed Cyberlake that allows you to store UNLIMITED 3D assets with a global cdn just for a flat $50. Pre-signup for access -- https://lnkd.in/gNGeyDvi #cloud #mixedreality

13

Some thoughts on why I don't see AI replacing FPV Pilots... Yet. Adaptability and Problem-Solving: FPV drones often operate in environments that require split-second decisions, rapid adjustments, and an understanding of nuanced physical cues. AI lacks the ability to intuitively "feel" these dynamics in unpredictable situations. Creative Judgment: No AI can currently match the creative intuition of a skilled FPV pilot. Cinematic FPV flying involves an artistic understanding of movement, framing, and timing, which is inherently human. Complex Environments: While autonomous systems excel in open, structured settings, FPV drones are often used in cluttered, unstructured environments. Pilots navigate these spaces with the kind of precision and intuition that AI struggles to replicate. #uas #pt107 #drones #fpv https://lnkd.in/giUmFQB7

1

Team Tumbleweed Software Engineering Team will be presenting our findings in a paper about Location and Attitude Determination (LAD) on Mars at IAC2024.   The paper describes how using a swarm of rovers enhances data collection efficiency and explains the critical role of a LAD algorithm in supporting scientific objectives, as well as elaborating on the challenges associated with a LAD algorithm tailored for a lightweight, wind-driven, rolling, low-cost Mars rover.   Classification of sensor mass and size is usually subjective. My contribution is proposing absolute values for the classification of sensors. The advantages and disadvantages of each sensor type in terms of accuracy, mass, size and computational processing requirement are also explained and compared. Come check out our presentation tomorrow at Orange Hall 2 at 10:51! Tim Holthuijsen, William Moretti, Abhimanyu K, Marcus Dyhr, Julian Rothenbuchner

17

Here is the last video (for now?) in the series focusing on common MBSE diagrams. This one covers State Machine Diagrams! I want to extend a big thanks to everyone who helped me make these videos, but especially Dan Mullen, Josh Kahn, Govind N Malleichervu, and Brian Douglas! What MBSE topics would you like to learn about next? What are some challenges you are facing? Link to the series in the comments below. #MBSE #DigitalEngineering #SystemEngineering

75

2 Comments

The sky is no longer the limit for #autonomous tech. With the unmanned aerial vehicle (UAV) market set to grow to $48.5B by 2029, aerospace and defense are at the edge of a new era. See what's driving this rapid growth. https://bit.ly/3NIRsvM

1

The sky is no longer the limit for #autonomous tech. With the unmanned aerial vehicle (UAV) market set to grow to $48.5B by 2029, aerospace and defense are at the edge of a new era. See what's driving this rapid growth. https://bit.ly/40oieRu

1

The sky is no longer the limit for #autonomous tech. With the unmanned aerial vehicle (UAV) market set to grow to $48.5B by 2029, aerospace and defense are at the edge of a new era. See what's driving this rapid growth. https://bit.ly/4hDai5k

1