![]() I had initially given them the task to estimate position by tracking the length of the cycloid that is made when the wheel moves several revolutions. The embedded systems team at AGV group designed a velocity sensor for our Mahindra E2O vehicle as the inbuilt encoder on the car has a very low resolution of 1kmph. Posted in AGV, Blog, Projects, Robotics, Software Awesome makeshift wheel mountable velocity sensor I really feel I should be able to complete it by the end of this week. Once it’s here, I guess it won’t be too long a job. With the control systems research, the motor still remains doubtful, and hence we have to wait for the repaired motor. The overall net was quite fast (more than 100 tests per second) and was generalizing well. But in the end, I managed to get a 99.2% test case accuracy on the test set by the fifth epoch. I had some issues with CUDA versions and forwarding the image as a CudaTensor. The input is a 256x256x3 image (3 is the channel count) I’d add the screenshot of the net and it’s result on a full size image by tomorrow. Although I was specifically using Torch, the overall experience of using the tool was fantastic and saved me a lot of calculation time. It lets you add Convolution, Pooling, Linear, ReLU, Fully connected and Output layers using a GUI and finally rolls out a TensorFlow code for the same. I finally designed a network using an awesome tool I recently discovered, named ConvNet Designer. Sample Positive from the dataset Sample Negative from the dataset I studied few resources where similar classification was done, and I wanted my net to be a bit smaller than AlexNet because the classification was not THAT difficult. Meanwhile, I also took some time out to finally complete the ConvNet and the code for classifying segments of a video live stream as bumpers and background. Talk about having the quietest Diwali (the festival of lights in India), the entire evening of mine was spent on the paper(s?) I’m working on, updating my resume and searching for procurements. With the new Husky UGV about to arrive and the work on the DuneBuggy almost ready for test, I anticipate awesomeness ahead! Posted in AGV, Blog, Hardware, Projects, Robotics Bumper Detection ConvNet: Results SICK LiDaR scanĪlso, we installed and ran Autoware on the AGV PC and ran the inbuilt SLAM module using the Velodyne, and the results were impressive. I had fun interfacing these and testing out the raw data. Coloured PointCloud from MultisenseĪlso a few new sensors have arrived in the lab, like the shiny new SICK LMS511 LiDaR and the Novatel GPS sensor. I’d try to implement robust bumper detection and resulting creep and crawl control soon and am also working on stereo visual odometry. As a result, I now am working primarily with the Carnegie Multisense S7 stereo camera. I have resumed work with AGV, now working on sensors more commonly found on real world cars due to the ongoing Rise Prize project. I haven’t been posting for a while, with the university registrations and stabilizing into the monotony of the life taking some time.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |