Multi-Task Learning with Sequence-Conditioned Transporter Networks
This work addresses the problem of enabling robots to solve multiple manipulation tasks for industrial applications, representing an incremental advancement through hybrid methods.
The paper tackles the challenge of scaling learning-based approaches for multi-task robot manipulation by introducing a new benchmark suite, MultiRavens, and a vision-based system, Sequence-Conditioned Transporter Networks, which significantly improves pick-and-place performance on novel 10 multi-task benchmark problems and enhances learning on individual tasks.
Enabling robots to solve multiple manipulation tasks has a wide range of industrial applications. While learning-based approaches enjoy flexibility and generalizability, scaling these approaches to solve such compositional tasks remains a challenge. In this work, we aim to solve multi-task learning through the lens of sequence-conditioning and weighted sampling. First, we propose a new suite of benchmark specifically aimed at compositional tasks, MultiRavens, which allows defining custom task combinations through task modules that are inspired by industrial tasks and exemplify the difficulties in vision-based learning and planning methods. Second, we propose a vision-based end-to-end system architecture, Sequence-Conditioned Transporter Networks, which augments Goal-Conditioned Transporter Networks with sequence-conditioning and weighted sampling and can efficiently learn to solve multi-task long horizon problems. Our analysis suggests that not only the new framework significantly improves pick-and-place performance on novel 10 multi-task benchmark problems, but also the multi-task learning with weighted sampling can vastly improve learning and agent performances on individual tasks.