Kristin Sallai’s Post

Sonair has raised $6.8 million in funding from early-stage specialists Skyfall Ventures and RunwayFBU, along with earlier investors. Ultrasound is perhaps best known as the technology that enables noninvasive body scans and underwater communication and can help us park our cars. Sonair out of Norway wants to employ it for something else: 3D computer vision used in autonomous hardware applications. Sonair believes the company’s application of ultrasound technology — a groundbreaking approach that reads sound waves to detect people and objects in 3D, with minimal energy and computational requirements — can be the basis of more useful and considerably less expensive solutions than today’s more standard approach using lidar. https://lnkd.in/eK-XYxgF

I have recently been working on an autonomous robot which has LIDAR. Now, I want to incorporate the Intel RealSense D455 for RGB-D SLAM. What are my best options for a low computational cost SLAM? Currently, I am considering the ORB-SLAM v3. Note that I am restricted to using the Jetson Orin Nano and not some high-end GPU for this task. #robotics #autonomous #SLAM #IntelRealSense #LIDAR

Though this article took place back in June, it's far too interesting not to share! MIT is using lidar and shadows to detect objects that cannot be seen by the sensor. This technology could significantly enhance computer vision in the future across a range of lidar applications. #technology #innovation #sensors #lidar #computervision

🤖✨ Embodied Neuromorphic Artificial Intelligence for Robotics Neuromorphic AI in robotics is moving towards creating robots that can interact, respond, and adapt to their environment much like a human brain does. The components consists of: 👁️ Sensors: Robots are equipped with event-based sensors like Dynamic Vision Sensors (DVS) and a suite of conventional sensors including GPS, LiDAR, and IMU, blending rapid data acquisition with precise positioning. 🧠 Computation: At the core lies Spiking Neural Networks (SNNs) that process these inputs. Software and hardware considerations include neural encoding, event-based data conversion, and mapping models on energy-efficient neuromorphic hardware. 🔩 Actuation: A critical piece of the puzzle, ensuring robots can take actions. The system must account for application constraints, whether it’s an Unmanned Aerial Vehicle (UAV) or an Industrial Assembly Robot. A lot of work remains such as accuracy and energy efficiency, developing benchmarks, improving reliability, security and more. Abs: https://lnkd.in/gETvgcCW #AI #GENAI #LLM #DEEPLLEARNING #Robotics #NeuromorphicAI #FutureOfAI

🚀 𝑰𝒏𝒕𝒆𝒈𝒓𝒂𝒕𝒊𝒏𝒈 2𝑫 𝑳𝒊𝑫𝑨𝑹 𝒇𝒐𝒓 𝑬𝒏𝒉𝒂𝒏𝒄𝒆𝒅 𝑹𝒐𝒃𝒐𝒕 𝑵𝒂𝒗𝒊𝒈𝒂𝒕𝒊𝒐𝒏🚀 Integrating 2𝘋 𝘓𝘪𝘋𝘈𝘙 with my mobile robot for real-time imaging and environment mapping. The LiDAR enables the robot to measure distances to surrounding objects and create a 2D map of its environment. This is a critical step in improving the robot's navigation, obstacle detection, and path planning capabilities. Using ROS 2, the data from the LiDAR will help the robot make informed decisions in real time, bringing us closer to fully autonomous operation. 𝘐𝘵'𝘴 𝘧𝘢𝘴𝘤𝘪𝘯𝘢𝘵𝘪𝘯𝘨 𝘵𝘰 𝘸𝘪𝘵𝘯𝘦𝘴𝘴 𝘩𝘰𝘸 𝘴𝘦𝘯𝘴𝘰𝘳 𝘧𝘶𝘴𝘪𝘰𝘯 𝘤𝘢𝘯 𝘣𝘳𝘪𝘯𝘨 𝘴𝘮𝘢𝘳𝘵𝘦𝘳 𝘳𝘰𝘣𝘰𝘵𝘪𝘤 𝘴𝘺𝘴𝘵𝘦𝘮𝘴 𝘵𝘰 𝘭𝘪𝘧𝘦! #Robotics #LiDAR #AutonomousSystems #MobileRobots #ROS2 #SensorFusion #EnvironmentMapping #PathPlanning #Navigation #Mechatronics #RobotImaging #RealTimeData #EngineeringInnovation #TechExploration #Automation

Interesting to know #cloud #projectmanagement #projects #future #news #television #ai #artificialintelligence #technology #technologies #electronics #manager #managers #ceo #education #university #bigdata #datacenter #business #company #java #SQL #datascience #machinelearning #deaplearning #ki #deaplearning #azur #PTTglobal #PTTtechnology #excel #newyear #newyearresolution #ExcelTraining #DataAnalysis #PowerQuery #PowerPivot #ExcelMastery #2025Ready #ProfessionalDevelopment #charisma #leadershipdevelopment #executivecoaching #lifecoaching #highimpact #entrepreneurship #personaldevelopment #emotionalintelligence #softwareengineer #softwareengineers #django #samurai #irobot #nobel

As noted in this article, the collaboration between Luminar LiDAR and Applied Intuition's toolchain is indeed powerful. It’s interesting to observe how different companies are leveraging these technologies. #lidar #devops #autonomous https://lnkd.in/gSGDfw3Z

🚘 UK startup Phlux Technology is aiming for the automotive LiDaR market with a new type of sensor as it looks to raise funds. 💡 👉 The Aura 1550 nm avalanche photo diodes (APDs) developed by Phlux were launched in January 2024 and are 12X more sensitive than other best-in-class InGaAs APDs. 👉This means that the operating range of IR-based systems can be immediately extended by up to 50% as well as offering accuracy and environmental stability as a drop in replacement. 💬 Benjamin White, CEO at Phlox, said: “For us its about scaling. We are selling into low volume, high value applications and we plan to scale for the automotive markets to millions of units. “We’ve got all the foundations in place with semiconductor supply chains and a number of foundries for the III-V IR sensor, and most of it is in Europe and the US. The packaging depends on the applications.” Read more here: https://lnkd.in/gqpCTN7x #Automotiveinnovation  #semiconductor  #automotivetechnology  #lidar

In today's rapidly evolving field of computer vision, even Tesla is starting to recognize the potential of lidar systems, a significant shift from their previous stance on the technology's effectiveness and cost. #sensors #lidar #computervision #technology #manufacturing #robotics

Having worked extensively with both Lidar and camera-based SLAM (Simultaneous Localization and Mapping) systems, I am left wondering about one very important question: which sensor will be the key to the future of autonomous systems? Robotics aficionados have long favored Lidar SLAM because of its accurate 3D spatial data. However, the extensive application of Lidar sensors has been limited by their high cost and complexity. What obstacles need to be removed in order to increase Lidar's accessibility? However, a more affordable option is provided by camera-based Visual SLAM. Computer vision advances have made it possible for cameras to give rich visual data for localization and mapping. Nevertheless, problems like glare, dim lighting, and the requirement for surroundings with plenty of features continue to exist. In what ways might these obstacles be overcome to improve the dependability of camera-based systems? Without a doubt, the trend is moving in the direction of camera-based methods because of the accessibility of affordable, high-quality cameras and the developments in deep learning. Particularly promising have been learning-based methods, which frequently achieve 99% accuracy rates. What about the 1% failure rate, though? In the face of uncertainty, how can the security and dependability of autonomous systems be guaranteed? Ultimately, both camera-based SLAM and Lidar will have an impact on how robotics and autonomous systems develop in the future. But it's imperative that we face the difficulties and unknowns that lie ahead as we make our way across this terrain. I therefore put these questions to you: which sensor do you believe is the key to the future, and how can we get beyond the obstacles in the way of securing a trustworthy and safe autonomous future? I would be delighted to hear about and establish connections with anybody who are developing any novel learning-based approaches in this area and in this direction. #autonomoustech #innovationjourney #sensorfusion #CameraPerception #AutomotiveTech #InnovationDriven #bgsw #bosch #AutonomousSystems #AI #MachineLearning #Robotics #Automation #IoT #ArtificialIntelligence #nvidia #nvidiaresearch #qualcomm #SmartTechnology #AutonomousVehicles #Innovation #TechTrends #FutureOfWork #Engineering #AutonomousDrones #SelfDrivingCars #SmartCities #MachineVision #Robotics #ros #ros2 #osrf #jderobot #analogdevices #texasinstruments #tatatechnologies #IoT #AutonomousDrones #SelfDrivingCar #ComputerVision #SLAM #localisation #mapping