Task-oriented grasping (TOG) refers to the problem of predicting grasps on an object that enable subsequent manipulation tasks. To model the complex relationships between objects, tasks, and grasps, existing methods incorporate semantic knowledge as priors into TOG pipelines. However, the existing semantic knowledge is typically constructed based on closed-world concept sets, restraining the generalization to novel concepts out of the pre-defined sets. To address this issue, we propose GraspGPT, a large language model (LLM) based TOG framework that leverages the open-end semantic knowledge from an LLM to achieve zero-shot generalization to novel concepts. We conduct experiments on Language Augmented TaskGrasp (LA-TaskGrasp) dataset and demonstrate that GraspGPT outperforms existing TOG methods on different held-out settings when generalizing to novel concepts out of the training set. The effectiveness of GraspGPT is further validated in real-robot experiments. Our code, data, appendix, and video are publicly available at https://sites.google.com/view/graspgpt/.

The prospect of assistive robots aiding in object organization has always been compelling. In an image-goal setting, the robot rearranges the current scene to match the single image captured from the goal scene. The key to an image-goal rearrangement system is estimating the desired placement pose of each object based on the single goal image and observations from the current scene. In order to establish sufficient associations for accurate estimation, the system should observe an object from a viewpoint similar to that in the goal image. Existing image-goal rearrangement systems, due to their reliance on a fixed viewpoint for perception, often require redundant manipulations to randomly adjust an object's pose for a better perspective. Addressing this inefficiency, we introduce a novel object rearrangement system that employs multi-view fusion. By observing the current scene from multiple viewpoints before manipulating objects, our approach can estimate a more accurate pose without redundant manipulation times. A standard visual localization pipeline at the object level is developed to capitalize on the advantages of multi-view observations. Simulation results demonstrate that the efficiency of our system outperforms existing single-view systems. The effectiveness of our system is further validated in a physical experiment.

To perform household tasks, assistive robots receive commands in the form of user language instructions for tool manipulation. The initial stage involves selecting the intended tool (i.e., object grounding) and grasping it in a task-oriented manner (i.e., task grounding). Nevertheless, prior researches on visual-language grasping (VLG) focus on object grounding, while disregarding the fine-grained impact of tasks on object grasping. Task-incompatible grasping of a tool will inevitably limit the success of subsequent manipulation steps. Motivated by this problem, this paper proposes GraspCLIP, which addresses the challenge of task grounding in addition to object grounding to enable task-oriented grasp prediction with visual-language inputs. Evaluation on a custom dataset demonstrates that GraspCLIP achieves superior performance over established baselines with object grounding only. The effectiveness of the proposed method is further validated on an assistive robotic arm platform for grasping previously unseen kitchen tools given the task specification. Our presentation video is available at: https://www.youtube.com/watch?v=e1wfYQPeAXU.

Generalization is one of the most important problems in deep learning (DL). In the overparameterized regime in neural networks, there exist many low-loss solutions that fit the training data equally well. The key question is which solution is more generalizable. Empirical studies showed a strong correlation between flatness of the loss landscape at a solution and its generalizability, and stochastic gradient descent (SGD) is crucial in finding the flat solutions. To understand how SGD drives the learning system to flat solutions, we construct a simple model whose loss landscape has a continuous set of degenerate (or near degenerate) minima. By solving the Fokker-Planck equation of the underlying stochastic learning dynamics, we show that due to its strong anisotropy the SGD noise introduces an additional effective loss term that decreases with flatness and has an overall strength that increases with the learning rate and batch-to-batch variation. We find that the additional landscape-dependent SGD-loss breaks the degeneracy and serves as an effective regularization for finding flat solutions. Furthermore, a stronger SGD noise shortens the convergence time to the flat solutions. However, we identify an upper bound for the SGD noise beyond which the system fails to converge. Our results not only elucidate the role of SGD for generalization they may also have important implications for hyperparameter selection for learning efficiently without divergence.