Seg-CURL: Segmented Contrastive Unsupervised Reinforcement Learning for Sim-to-Real in Visual Robotic Manipulation
نویسندگان
چکیده
Training image-based reinforcement learning (RL) agents are sample-inefficient, limiting their effectiveness in real-world manipulation tasks. Sim2Real, which involves training simulations and transferring to the real world, effectively reduces dependence on data. However, performance of transferred agent degrades due visual difference between two environments. This research presents a low-cost segmentation-driven unsupervised RL framework (Seg-CURL) solve Sim2Real problem. We transform input RGB views proposed semantic segmentation-based canonical domain. Our method incorporates levels Sim2Real: task-level transfers observation-level simulated U-nets segment scenes. Specifically, we first train contrastive RL(CURL) with segmented images simulation environment. Next, employ U-Nets robotic hand-view side-view during robot control. These U-Net pre-trained synthetic segmentation masks environment fine-tuned only 20 images. evaluate robustness both Seg-CURL is robust texture, lighting, shadow, camera position gap. Finally, our algorithm tested Baxter dark cube lifting task success rate 16/20 zero-shot transfer.
منابع مشابه
Reinforcement Learning for Appearance Based Visual Servoing in Robotic Manipulation
The objective of this paper is to develop a new appearance based visual servoing method that needs no prior structuring of the environment and also eliminates the correspondence problem associated with conventional visual servoing methods. Detailed description of object appearance and its generation are provided in this paper. In addition, owing to the non-linear and high dimensional nature of ...
متن کاملComposable Deep Reinforcement Learning for Robotic Manipulation
Model-free deep reinforcement learning has been shown to exhibit good performance in domains ranging from video games to simulated robotic manipulation and locomotion. However, model-free methods are known to perform poorly when the interaction time with the environment is limited, as is the case for most real-world robotic tasks. In this paper, we study how maximum entropy policies trained usi...
متن کاملDeep Reinforcement Learning for Robotic Manipulation
Reinforcement learning holds the promise of enabling autonomous robots to learn large repertoires of behavioral skills with minimal human intervention. However, robotic applications of reinforcement learning often compromise the autonomy of the learning process in favor of achieving training times that are practical for real physical systems. This typically involves introducing hand-engineered ...
متن کاملFlexible Robotic Grasping with Sim-to-Real Transfer based Reinforcement Learning
Robotic manipulation requires a highly flexible and compliant system. Task-specific heuristics are usually not able to cope with the diversity of the world outside of specific assembly lines and cannot generalize well. Reinforcement learning methods provide a way to cope with uncertainty and allow robots to explore their action space to solve specific tasks. However, this comes at a cost of hig...
متن کاملReinforcement Learning for Robotic Locomotions
● Modifications on constraints Since TRPO is a constraint optimization problem, our first thought is replacing the KL constraint by some other constraints that also measure policy similarity. A natural thought would be using MSE loss on . We noticed later that this in fact corresponds to the standard policy gradient update. We have also tried to directly optimize the objective without any const...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: IEEE Access
سال: 2023
ISSN: ['2169-3536']
DOI: https://doi.org/10.1109/access.2023.3278208