# Weekly Brief of My Research (5th July'19) ***Objective*** - I am trying to automate the procedure of detecting the pose goal and making the robot arm move there and grasp that object. I am also trying to find out methods for Peg-in-hole assembly with high accuracy. ***Introduction*** - I have been reading various research papers in this regard and I found quite intresting research papaers. Researchers have used various methods to do that. I am mentioning the brief of those methods by which we can try to accomplish our task with high accuracy. * This method consists of an analysis of the state of contact between the peg and the hole as well as a strategy to overcome the inevitable positional uncertainty of the hole incurred in the recognition process. * In this method an autonomous alignment method by force/torque measurement before insertion phase has been proposed. With the estimated values, the robot can autonomously correct the misalignment. * This method introduces a framework for teaching robot peg-hole-insertion from human demonstration. This works on the feedback mechanism after demosntration is done and robot arm is on the way to insert peg in hole with required precision. * This method is based on deep reinforcement learning, a skill-acquisition method is used to solve the posed problems of uncertainty in a complex assembly process. They have performed simulations in Gazebo and also they have tried in the real world environment for assembly. * In this method, they have used a depth camera mounted near the robot’s end-effector. They present a high-accuracy two-stage pose estimation procedure based on deep convolutional neural networks, which includes detection, pose estimation, refinement, and handling of near-and full symmetries of parts. The networks are trained on simulated depth images withmeans to ensure successful transfer to the real robot. ***Plan for the next week -*** I will read some more research papaers in next week. By the next week I will try to know the best suitable procedure which we should follow for the preciese peg-in-hole assembly with our project. I will prepare a summary for the various methods which we can follow and will try to attempt a few of them. In the next week, I will try to automate the process of estimation of pose goal.