I previously shared my work using a custom YOLOv8 model for object detection on KITTI and Nuscenes datasets with LiDAR data projected onto a bird's-eye view (BEV).
Now, I have trained a custom Yolov8 model for front camera images and have done the visualizations of both the camera and LiDAR BEV model outputs! This allows for a fascinating comparison of how each sensor "sees" the environment. This can also help autonomous vehicles get a 3D view of the environment.
This combined visualization is a great step towards achieving my goal of using weighted fusion for the final results.
#ComputerVision#ObjectDetection#KITTI#YOLOv8#BirdsEyeView#DataScience#MachineLearning#Robotics
The client was looking for a partner to annotate dynamic objects from Camera and LIDAR data to create high-quality AI training data for autonomous vehicle perception systems.
RMSI annotated dynamic objects like vehicles, pedestrians, and animals in 2D camera images using 2D Bounding Box tool, and in 3D Lidar scenes using 3D Cuboid Bounding Box.
How did the client benefit?
1. The high-quality HD maps will assist the client in precise anticipation of various objects in the surrounding while creating an automated driving system.
2. With these training datasets, the autonomous car will be able to get familiar with an environment and will also keep a track of moving objects.
#lidar#3d#mapping#3dmaps#annotation#2dmaps#hd#hdmaps
The client was looking for a partner to annotate dynamic objects from Camera and LIDAR data to create high-quality AI training data for autonomous vehicle perception systems.
RMSI annotated dynamic objects like vehicles, pedestrians, and animals in 2D camera images using 2D Bounding Box tool, and in 3D Lidar scenes using 3D Cuboid Bounding Box.
How did the client benefit?
1. The high-quality HD maps will assist the client in precise anticipation of various objects in the surrounding while creating an automated driving system.
2. With these training datasets, the autonomous car will be able to get familiar with an environment and will also keep a track of moving objects.
#lidar#3d#mapping#3dmaps#annotation#2dmaps#hd#hdmaps
🌟 Celebrating a Major Breakthrough in Robotics! 🌟
We’re thrilled to share a sneak peek into our latest project, where we're testing the capabilities of a 2D LIDAR system mounted on a mobile platform. 📈🚀
In this video, you’ll see how our advanced LIDAR technology is performing real-time mapping as the small vehicle navigates through its environment. This is a crucial step in enhancing autonomous navigation and improving accuracy in dynamic settings. 🛰️🚗
Key highlights:
🔹 Real-time Data Processing: Observe how the LIDAR captures and processes data to create precise maps on the go.
🔹 Dynamic Environment Navigation: See the system's performance as it adapts to different obstacles and terrain.
🔹 Innovative Technology: Get a glimpse of how this technology could revolutionize autonomous systems and smart navigation solutions.
We’re incredibly proud of our team's dedication and innovation in pushing the boundaries of what’s possible in robotics and automation. Stay tuned for more updates as we continue to refine and expand our technology!
🔍 #Robotics#LIDAR#Innovation#Automation#TechDevelopment#AutonomousVehicles#Mapping
The technology of LiDAR (light detection and ranging) is common in autonomous vehicles, robotics, and many other applications for mapping three-dimensionally or detecting objects. Infosearch can help you with your diverse LiDAR annotation requirements.
https://lnkd.in/dCkin-h9
Read more about our LiDAR annotation services here.
https://lnkd.in/gEf7xc29
Contact us for your annotation requirements.
enquiries(@)infosearchbpo(dot)com
#3DLiDAR#annotationservices#3DModel#AutonomousVehicles#Robotics
🌐🚗 Did you know that one of the key components of autonomous vehicles is the LiDAR sensor? LiDAR, which stands for Light Detection and Ranging, scans the environment around the vehicle and creates a detailed 3D map. These real-time maps are critical for helping autonomous vehicles navigate safely. To effectively function, they require regular cleaning to remove any dust or dirt. A simple wipe with a clean microfiber cloth can do wonders for detection and accuracy. Keep your autonomous journey smoother and safe! 💡🚀 #AutonomousVehicles#TechTip#LiDAR#FutureTech 🖥️🧩
How to reverse into trailers autonomously? This is one of the final steps in completing fully autonomous yard operations.
Until now, it was very challenging for the terminal tractors to control the position and height of the kingpin as connection of the fifth wheel. For this purpose, Autonomous Solutions, Inc. (ASI) now uses a Vehicle Automation Kit (VAT) and 2D LiDAR sensors.
Learn how the sensors help detecting the trailer's kingpin in our blog: https://bit.ly/3S3QXiK#autonomousdriving#autonomousvehicles#SICKSensorIntelligence#sensors
🚗 Mastering Fully Autonomous Yard Operations: Reversing into Trailers Made Easy! 🚗
Something I can use on those "everything goes wrong" days. 😅
Ever wondered how terminal tractors effortlessly reverse into trailers autonomously? 🤖
Unlock the secret with @Autonomous Solutions, Inc. (ASI) as they unveil their groundbreaking approach!
🔧 The Challenge: Controlling kingpin position and height during the fifth wheel connection.
💡 The Solution: Enter the Vehicle Automation Kit (VAT) coupled with state-of-the-art 2D LiDAR sensors.
Join me on this educational journey to explore the pivotal role of technology in achieving seamless autonomy in yard operations.
Stay tuned for a glimpse into the future of logistics! 🚚
#AutonomousTech#LogisticsInnovation#TechBreakthrough
How to reverse into trailers autonomously? This is one of the final steps in completing fully autonomous yard operations.
Until now, it was very challenging for the terminal tractors to control the position and height of the kingpin as connection of the fifth wheel. For this purpose, Autonomous Solutions, Inc. (ASI) now uses a Vehicle Automation Kit (VAT) and 2D LiDAR sensors.
Learn how the sensors help detecting the trailer's kingpin in our blog: https://bit.ly/3S3QXiK#autonomousdriving#autonomousvehicles#SICKSensorIntelligence#sensors
🚀 Exciting Progress in Autonomous Navigation with LiDAR on Clearpath's Husky Platform! 🚀
I'm thrilled to share my latest advancements in autonomous navigation using LiDAR on Clearpath Robotics' Husky platform. Check out my video showcasing three different tests:
-Maneuvering past two chairs
-Avoiding a person (my boy Adam Speed-Andrews 😊) standing in front of the robot
-Handling dynamic obstacles with two people, one appearing in front as the robot progresses towards its goal
In the video, you’ll see how the Husky uses LiDAR and the Nav2 stack to detect and navigate around obstacles, with real-time feedback visualized in RViz. This robust LiDAR-based navigation serves as my ground truth as I work towards implementing vision-only-based navigation using a stereo camera.
Using LiDAR as a benchmark is ideal due to its accuracy and reliability, making it a perfect reference for refining vision-based navigation systems. Currently, wheel odometry is being used for localization. However, one of the challenges with wheel odometry is its susceptibility to drift and inaccuracies over time, especially on uneven terrain. To address this, I plan to implement visual odometry in the future to test its reliability and improve overall navigation performance.
Stay tuned for more updates as I continue to push the boundaries of autonomous navigation technology!
#AutonomousNavigation#LiDAR#Robotics#ClearpathRobotics#Husky#RViz#Nav2#Innovation#Technology#StereoCamera#RoboticsResearch
GenAI Dev| 🧑🏫Pursuing M.Tech, AI&ML@BITS PILANI |🧑💻Kaggle Master| 📃Certified: TensorFlow Dev, Nvidia Jetson AI specialist, Azure AI 900|⚙️Enthusiastic & Curious about AI (GNN, RL, 2D & 3D CV, Audio)
6moGreat to see this kind of insight Manish Nayak 🤗