Simulation is a cheap and abundant source of data, which allows training deep neural networks on huge tailor-made datasets. Especially the ease of data labeling – in particular for costly labels such as in pixel-wise segmentation – makes simulation an interesting tool in deep learning. However, training a network on simulated data with the intent […]
I have published a new blog showing my latest effort of training a position/pose estimator based purely on simulated data. Therefore, I have extended an approach considered as domain randomization. You can find the blog here. Check it out and leave a comment!
I have uploaded a new video showing how to use redundancy resolution in different policy search methods – including reinforcement learning and evolutions strategies – to embed secondary objectives without any reward shaping. This approach is evaluated in simulation and on the real robot. It focuses on the task of robotic manipulation. Here it is: […]