Multi-Object Graph Affordance Network: Goal-Oriented Planning through Learned Compound Object Affordances
This addresses a gap in robot learning for complex object interactions, though it is incremental as it extends existing affordance learning methods to compound objects.
The paper tackles the problem of learning affordances for compound objects composed of multiple objects, proposing a graph neural network model that predicts interaction outcomes and enables goal-oriented planning for tasks like stacking and inserting. The system successfully modeled affordances for concave and convex objects in simulated and real-world environments, with benchmarking showing advantages over a baseline.
Learning object affordances is an effective tool in the field of robot learning. While the data-driven models investigate affordances of single or paired objects, there is a gap in the exploration of affordances of compound objects composed of an arbitrary number of objects. We propose the Multi-Object Graph Affordance Network which models complex compound object affordances by learning the outcomes of robot actions that facilitate interactions between an object and a compound. Given the depth images of the objects, the object features are extracted via convolution operations and encoded in the nodes of graph neural networks. Graph convolution operations are used to encode the state of the compounds, which are used as input to decoders to predict the outcome of the object-compound interactions. After learning the compound object affordances, given different tasks, the learned outcome predictors are used to plan sequences of stack actions that involve stacking objects on top of each other, inserting smaller objects into larger containers and passing through ring-like objects through poles. We showed that our system successfully modeled the affordances of compound objects that include concave and convex objects, in both simulated and real-world environments. We benchmarked our system with a baseline model to highlight its advantages.