Behavior Cloning-Based Robot Active Object Detection With Automatically Generated Data and Revision Method

Behavior Cloning-Based Robot Active Object Detection With Automatically Generated Data and Revision Method
复制标题

基于行为克隆的机器人主动物体检测自动生成数据和修正方法

DOI:
10.1109/tro.2022.3191745
复制
发表时间:
2023-02
影响因子:
7.8
通讯作者:
Shuo Liu
Shuo Liu
中科院分区:
计算机科学1区
文献类型:
--
作者:
Shaopeng Liu;Guohui Tian;Xuyang Shao;Shuo Liu

文献摘要

参考文献

相似文献

主动物体检测(AOD)是机器人领域最大的挑战之一,也是本文的重点。目前大多数的AOD方法都是通过强化学习(RL)算法开发的,但它们可以在训练时间,训练效率,模型性能和模型预测方面进一步改进。因此,不同于现有的工作,我们提出了一种基于自动生成的数据训练的行为克隆的AOD方法。我们将AOD任务转化为动作分类问题,不仅缩短了训练时间,而且提高了训练效率和模型性能。由于没有可用的专家数据来训练所提出的基于分类的AOD模型,我们设计了一个自主的数据生成方法,以避免大量的手动注释。我们引入了一个多输入网络,以更好的避障和AOD性能,其中的深度图像被添加到帮助机器人感知环境和物体的距离信息。此外,我们还提出了一种模型预测的修正方法,以减少复合误差的积累,有效地提高了长路径AOD任务的成功率。通过对比实验和烧蚀实验,我们在一个AOD数据集上对我们的方法进行了广泛的评估,证明了我们的方法在AOD性能和效率上优于其他方法。此外,在现实世界的场景与TIAGo机器人的AOD实验表明,我们的方法的有效性。
Active object detection (AOD), one of the greatest challenges in the robotics field, is the main focus of this article. Most current AOD methods are developed by reinforcement learning (RL) algorithms while they can be further improved in the aspects of training time, training efficiency, model performance, and model prediction. Therefore, different from the existing works, we propose an AOD method based on behavior cloning trained by automatically generated data. We transform the AOD task into an action classification problem to not only shorten the training time but also improve the training efficiency and model performance. As there is no available expert data for training the presented classification-based AOD model, we design an autonomous method of data generation to avoid the large amounts of manual annotations. We introduce a multiinput network for better obstacle avoidance and AOD performance, where the depth image is added to help the robot to perceive distance information of environments and objects. Moreover, we develop a revision method for model prediction to reduce the accumulation of compounding error, which improves the successful rate of the long path AOD tasks effectively. We extensively evaluate our method on an AOD dataset by the comparable experiments and the ablation study, proving that our approach outperforms other methods in AOD performance and efficiency. In addition, the AOD experiments in the real-world scenario with a TIAGo robot indicate the validity of our method.
DOI: 10.1109/cvprw.2018.00277
发表时间: 2018-06
期刊: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
影响因子: --
作者:
Phil Ammirato;A. Berg;J. Kosecka
通讯作者: Phil Ammirato;A. Berg;J. Kosecka
DOI: 10.1109/iros40897.2019.8967805
发表时间: 2019-11
期刊: 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
影响因子: --
作者:
Jan Fabian Schmid;M. Lauri;S. Frintrop
通讯作者: Jan Fabian Schmid;M. Lauri;S. Frintrop
DOI: 10.1007/s11263-019-01236-7
发表时间: 2017-02
影响因子: 19.5
作者:
Saurabh Gupta;Varun Tolani;James Davidson;S. Levine;R. Sukthankar;J. Malik
通讯作者: Saurabh Gupta;Varun Tolani;James Davidson;S. Levine;R. Sukthankar;J. Malik
DOI: --
发表时间: 2015-11
期刊: CoRR
影响因子: --
作者:
T. Schaul;John Quan;Ioannis Antonoglou;David Silver
通讯作者: T. Schaul;John Quan;Ioannis Antonoglou;David Silver
DOI: --
发表时间: 2016-02
期刊: ArXiv
影响因子: --
作者:
F. Iandola;Matthew W. Moskewicz;Khalid Ashraf;Song Han;W. Dally;K. Keutzer
通讯作者: F. Iandola;Matthew W. Moskewicz;Khalid Ashraf;Song Han;W. Dally;K. Keutzer