Objects rarely sit in isolation in everyday human environments. If we want robots to operate and perform tasks in our human environments, they must understand how the objects they manipulate will interact with structural elements of the environment for all but the simplest of tasks. As such, we'd like our robots to reason about how multiple objects and environmental elements relate to one another and how those relations may change as the robot interacts with the world. We examine the problem of predicting inter-object and object-environment relations between previously unseen objects and novel environments purely from partial-view point clouds. Our approach enables robots to plan and execute sequences to complete multi-object manipulation tasks defined from logical relations. This removes the burden of providing explicit, continuous object states as goals to the robot. We explore several different neural network architectures for this task. We find the best performing model to be a novel transformer-based neural network that both predicts object-environment relations and learns a latent-space dynamics function. We achieve reliable sim-to-real transfer without any fine-tuning. Our experiments show that our model understands how changes in observed environmental geometry relate to semantic relations between objects. We show more videos on our website: https://sites.google.com/view/erelationaldynamics.

Robots operating in human environments must be able to rearrange objects into semantically-meaningful configurations, even if these objects are previously unseen. In this work, we focus on the problem of building physically-valid structures without step-by-step instructions. We propose StructDiffusion, which combines a diffusion model and an object-centric transformer to construct structures given partial-view point clouds and high-level language goals, such as "set the table". Our method can perform multiple challenging language-conditioned multi-step 3D planning tasks using one model. StructDiffusion even improves the success rate of assembling physically-valid structures out of unseen objects by on average 16% over an existing multi-modal transformer model trained on specific structures. We show experiments on held-out objects in both simulation and on real-world rearrangement tasks. Importantly, we show how integrating both a diffusion model and a collision-discriminator model allows for improved generalization over other methods when rearranging previously-unseen objects. For videos and additional results, see our website: https://structdiffusion.github.io/.

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.

Geometric organization of objects into semantically meaningful arrangements pervades the built world. As such, assistive robots operating in warehouses, offices, and homes would greatly benefit from the ability to recognize and rearrange objects into these semantically meaningful structures. To be useful, these robots must contend with previously unseen objects and receive instructions without significant programming. While previous works have examined recognizing pairwise semantic relations and sequential manipulation to change these simple relations none have shown the ability to arrange objects into complex structures such as circles or table settings. To address this problem we propose a novel transformer-based neural network, StructFormer, which takes as input a partial-view point cloud of the current object arrangement and a structured language command encoding the desired object configuration. We show through rigorous experiments that StructFormer enables a physical robot to rearrange novel objects into semantically meaningful structures with multi-object relational constraints inferred from the language command.

Abstract-- Requiring multiple demonstrations of a task plan presents a burden to end-users of robots. However, robustly executing tasks plans from a single end-user demonstration is an ongoing challenge in robotics. We address the problem of one-shot task execution, in which a robot must generalize a single demonstration or prototypical example of a task plan to a new execution environment. Our experimental evaluations show that our knowledge representation makes more relevant generalizations that result in significantly higher success rates over tested baselines. The task generalization module incrementally generalizes platform, which resulted in the successful generalization of failed task plans by leveraging the learned knowledge graph to initial task plans to 38 of 50 execution environments with errors infer plan constituents (see Sec IV).

Prior work has shown that the multi-relational embedding objective can be reformulated to learn dynamic knowledge graphs, enabling incremental class learning. The core contribution of our work is Incremental Semantic Initialization, which enables the multi-relational embedding parameters for a novel concept to be initialized in relation to previously learned embeddings of semantically similar concepts. We present three variants of our approach: Entity Similarity Initialization, Relational Similarity Initialization, and Hybrid Similarity Initialization, that reason about entities, relations between entities, or both, respectively. When evaluated on the mined AI2Thor dataset, our experiments show that incremental semantic initialization improves immediate query performance by 21.3 MRR* percentage points, on average. Additionally, the best performing proposed method reduced the number of epochs required to approach joint-learning performance by 57.4\% on average.

The knowledge base completion problem is the problem of inferring missing information from existing facts in knowledge bases. Path-ranking based methods use sequences of relations as general patterns of paths for prediction. However, these patterns usually lack accuracy because they are generic and can often apply to widely varying scenarios. We leverage type hierarchies of entities to create a new class of path patterns that are both discriminative and generalizable. Then we propose an attention-based RNN model, which can be trained end-to-end, to discover the new path patterns most suitable for the data. Experiments conducted on two benchmark knowledge base completion datasets demonstrate that the proposed model outperforms existing methods by a statistically significant margin. Our quantitative analysis of the path patterns shows that they balance between generalization and discrimination.

General-purpose robots operating in a variety of environments, such as homes or hospitals, require a way to integrate abstract knowledge that is generalizable across domains with local, domain-specific observations. In this work, we examine different types and sources of data, with the goal of understanding how locally observed data and abstract knowledge might be fused.We introduce the Situated Robot Knowledge (SiRoK) framework that integrates probabilistic abstract knowledge and semantic memory of the local environment. In a series of robot and simulation experiments we examine the tradeoffs in the reliability and generalization of both data sources. Our robot experiments show that the variability of object properties and locations in our knowledge base is indicative of the time it takes to generalize a concept and its validity in the real world. The results of our simulations back that of our robot experiments, and give us insights into which source of knowledge to use for 31 types of object classes that exist in the real world.