Denis Marcon’s Post

Our first test platform for ADAS SLAM and Perception subsystems. ☺️ There are control algorithms but no actuators for now. 360 degree computation will be added after approval of the SLAM and Perception algorithms. #ADAS #ai #computervision #photogrammetry #autonomousdriving #autonomousvehicles #automotive #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

When it comes to Autonomous Vehicles and #ADAS (Advanced Driver Assistance Systems), precise LiDAR sensor simulations are critical. With the latest advancements in optical system design, our simulations now incorporate advanced surface-level optical properties, enhancing accuracy for optimal sensor performance and seamless integration. Ansys Optical provides a comprehensive multi-scale, multiphysics solution for researchers and engineers to design, analyze, and optimize all types of photonics and optics devices. 🔗 More about the 2024R2 software release: https://lnkd.in/dj-_KpXt #AutonomousVehicles #LiDAR #OpticalSystemDesign #Simulation

In our recent work titled, "SRFDet3D: Sparse Region Fusion based 3D Object Detection," we proposed a sparse-prior LiDAR-Camera fusion-based 3D object detection network for autonomous driving. ✨ Designed a novel proposal generation module to generate sparse 3D object proposals which are input-dependent (on both sensor modalities). ✨ Fusion of only sparse set of proposals to reduce computation cost. ✨ Comprehensive experiments on 3 large-scale nuScenes, KITTI and Waymo autonomous driving datasets. ✨ Robust in adverse weather conditions, such as rainy and night. ✨ Accepted for publication in Neurocomputing journal (IF=6.0). 📰 Paper: https://lnkd.in/dQSasY4J 👨💻 Code: https://lnkd.in/duHFBBia #ComputerVision #AutonomousDriving #AutonomousVehicles #MachineLearning #3DObjectDetection #Camera #LiDAR #SensorFusion #DeepLearning #KITTI #Waymo #nuScenes

Advancing LIDAR Technology: The Role of Multi-Gate FET Amplifiers           Designing amplifiers is crucial for advancing LIDAR technology, particularly in the context of self-driving vehicles. Conventional FETs face limitations in power, speed, and efficiency, prompting the development of multi-gate FETs to address these constraints.   An amplifier enhances the power level of an input signal, finding utility in various domains like audio systems, radar, and LIDAR technology. Among amplifier types, field-effect transistor (FET) amplifiers, particularly multi-gate FETs, stand out for their ability to finely control current flow, thus enhancing performance.   LIDAR (Light Detection and Ranging) technology, akin to RADAR but employing laser light, is pivotal in creating 3D object representations. Its applications span autonomous vehicles, surveying, and industrial automation, demanding considerations such as power efficiency, compactness, resolution, signal fidelity, and cost-effectiveness.   The design process for multi-gate FET amplifiers typically involves defining requirements, selecting suitable FET devices, crafting matching networks and bias circuits, optimizing via simulations, and prototyping.   Key design parameters for LIDAR systems encompass power efficiency, compactness, resolution, signal fidelity, and cost. Ensuring accuracy and reliability is paramount for the safety of autonomous vehicles and other applications reliant on LIDAR data. Moreover, managing costs is essential for technology accessibility and advancement.   Multi-gate FET amplifiers offer advantages like high power density, efficiency, and reduced power consumption, making them ideal for space-constrained applications like LIDAR. However, they pose challenges in terms of complexity and higher costs compared to conventional FETs.   Pipeloluwa Olayiwola hashtag #100daysampdesign hashtag #AnalogICDesign IEEE Solid-State Circuits Society

😤 𝐀𝐫𝐞 𝐲𝐨𝐮 𝐬𝐭𝐫𝐮𝐠𝐠𝐥𝐢𝐧𝐠 𝐭𝐨 𝐝𝐞𝐯𝐞𝐥𝐨𝐩 𝐚𝐧𝐝 𝐬𝐲𝐧𝐜𝐡𝐫𝐨𝐧𝐢𝐳𝐞 𝐲𝐨𝐮𝐫 𝐦𝐮𝐥𝐭𝐢-𝐬𝐞𝐧𝐬𝐨𝐫 𝐚𝐩𝐩𝐥𝐢𝐜𝐚𝐭𝐢𝐨𝐧𝐬? 😓 Getting a lidar, IMU, cameras and more sensors all together with perception algorithms can become really complex and require a lot of efforts. ⏱️ See how RTMaps middleware can help you to accelerate and ease your development of your multisensor applications, all synchronized in just 5 minutes! Period! Ready? Set.. Go! 💨 #RTMaps #middleware #sensors #lidar #camera #imu #software #softwaredevelopment #sdv #autonomous #ADAS #automateddriving #synchronization #data

Good morning I started to continue working on the embedded project this morning, I already have two cars and this afternoon I will get the third one and I will also attach a lidar sensor, then I will do a car race with all three. 🚗 🚗 🚗 #embedded #automotive #software #lidar #microcontrollers

One of the modalities vehicles use to sense their environment is radar. It provides good range and velocity information, but doesn't provide an image like a camera, (LiDAR is somewhere in between). Radar makers are working on the next gen, called 'Imaging Radar'. It is a good step forward, though still not with the angular resolution of LiDAR, or camera. To achieve this higher resolution the radar needs to have a larger aperture (the antenna array is larger), which means it takes longer for all the wavefronts from each antenna element to coalesce into a flat field. This is important so the 'picture' is not 'blurry'. The distance at which the field is flat enough is called the Far Field (FF). I have seen Imaging Radar modules where the FF is > 12 meters. Testing these modules in a controlled environment requires either a large anechoic chamber, or a curved reflector to flatten the field. This link provides a description (and some of the math) on how such a reflector works. https://lnkd.in/gsFVB-yY #ADAS, #AutomotiveRadar, #AutonomousVehicles, #Keysight

Wired Jul 2024 and Telsa Oct 2024: Shifting from #LiDAR to cameras. Current AVs rely heavily on LiDAR (Light Detection and Ranging) sensors, which use lasers to create a #3D #map of the surroundings. However, the tide is now turning towards #autofocus #cameras as the #future of #automotive #vision systems....#Automated & #Robotized #Camera #test & #validation #Labs will rise to the next level to ensure this type of progress ?

#Ultrasound is perhaps best known as the technology that enables noninvasive body scans and underwater communication and can help us park our cars. A young startup called #Sonair out of Norway wants to employ it for something else: #3Dcomputervision used in #autonomoushardware applications. Sonair’s founder and CEO Knut Sandven believes the company’s application of #ultrasoundtechnology — 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 #lessexpensive solutions than today’s more standard approach using #lidar. #ultrasoundtechnology #3Dcomputervision #autonomoushardware #sensors #beamforming