Office7 Trajectory

Dataset of the CoRL 2021 Paper "Self-Improving Semantic Perception for Indoor Localisation" Abstract: We propose a novel robotic system that can improve its perception during deployment. Contrary to the established approach of learning semantics from large datasets and deploying fixed models, we propose a framework in which semantic models are continuously updated on the robot to adapt to the deployment environments. By combining continual learning with self-supervision, our robotic system learns online during deployment without external supervision. We conduct real-world experiments with robots localising in 3D floorplans. Our experiments show how the robot's semantic perception improves during deployment and how this translates into improved localisation, even across drastically different environments. We further study the risk of catastrophic forgetting that such a continuous learning setting poses. We find memory replay an effective measure to reduce forgetting and show how the robotic system can improve even when switching between different environments. On average, our system improves by 60% in segmentation and 10% in localisation accuracy compared to deployment of a fixed model, and it maintains this improvement while adapting to further environments.CoRL 2021收录论文《面向室内定位的自改进语义感知（Self-Improving Semantic Perception for Indoor Localisation）》的数据集配套摘要：我们提出了一种全新的机器人系统，可在部署过程中持续优化自身感知能力。与现有从大规模数据集中学习语义并部署固定模型的经典范式不同，我们提出了一种可在机器人端持续更新语义模型以适配部署环境的框架。通过结合持续学习（continual learning）与自监督学习（self-supervision），该机器人系统可在部署过程中无需外部监督即可实现在线学习。我们使用在三维平面图中开展定位任务的机器人完成了真实世界实验。实验结果表明，即便在截然不同的环境中部署，机器人的语义感知能力在部署过程中得以提升，且该性能提升可转化为定位精度的优化。我们进一步研究了此类持续学习场景所面临的灾难性遗忘（catastrophic forgetting）风险，发现经验回放（memory replay）是缓解遗忘的有效手段，并证实即便在不同环境间切换时，该机器人系统仍可实现性能提升。与部署固定模型的方案相比，我们的系统平均在语义分割任务上提升60%，在定位精度上提升10%，且在适配更多新环境的过程中仍能维持该性能增益。