Abstract
Large-scale robotic policies trained on data from diverse tasks and robotic platforms hold great promise for enabling general-purpose robots; however, reliable generalization to new environment conditions remains a major challenge. Toward addressing this challenge, we propose a novel approach for uncertainty-aware deployment of pre-trained language-conditioned imitation learning agents. Specifically, we use temperature scaling to calibrate these models and exploit the calibrated model to make uncertainty-aware decisions by aggregating the local information of candidate actions. We implement our approach in simulation using three such pre-trained models, and showcase its potential to significantly enhance task completion rates. The accompanying code is accessible at the link: this https URL
Abstract (translated)
大规模机器人策略通过从多样任务和机器人平台的数据进行训练,具有很大的潜力,使通用机器人具有更大的潜力。然而,将这些策略应用到新环境条件仍然是一个主要挑战。为了解决这个挑战,我们提出了一个名为 uncertainty-aware deployment of pre-trained language-conditioned imitation learning agents 的全新方法。具体来说,我们使用温度缩放来校准这些模型,并通过聚合候选动作的局部信息来做出不确定性明智的决策。我们在仿真中使用三个这样的预训练模型来实现我们的方法,并展示其对显著提高任务完成率的可能性。附录中的代码可在此链接访问:https://
URL
https://arxiv.org/abs/2403.18222