CAREER: Active Bayesian Inference for Collaborative Robot Mapping

职业：协作机器人绘图的主动贝叶斯推理

• 批准号: 2045945
• 负责人: Nikolay Atanasov
• 金额: $ 60万
• 依托单位: University of California-San Diego
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-04-01 至 2026-03-31
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2045945&HistoricalAwards=false
• 关键词: CAREER; Active; Bayesian; Inference; Collaborative

Artificial perception techniques, allowing robot systems to know their location and surroundings using sensory data, have been instrumental for enabling robot automation outside of carefully controlled manufacturing settings. Current robot systems, however, remain passive in their perception of the world. Unlike biological systems, robots lack curiosity mechanisms for exploration and uncertainty mitigation, which are critical for intelligent decision making. Such capabilities are very important in disaster response, security and surveillance, and environmental monitoring, where it is necessary to quickly gain situational awareness of the terrain, buildings, and humans in the environment. The methods developed in this project will impact the design of mapping and active sensing algorithms for autonomous robot teams and their use in the aforementioned applications. This Faculty Early Career Development (CAREER) Program research develops fundamental robot autonomy capabilities that will also impact other domains relying on autonomous robots. In addition, the project will develop a suite of open-source education materials, including theoretical problems, projects, lectures, and exemplary implementations of core robotics algorithms, unified in an easily accessible simulation environment. This platform will support curriculum development for graduate students, as well as outreach and research-initiation activities for undergraduate and K-12 students.The research agenda will be achieved through two key technical innovations. First, the project will formally define an Active Bayesian Inference problem, seeking optimal control of sensing systems for minimum uncertainty estimation. Methods for distributed approximate dynamic programming that utilize the structure of the problem, induced by the functions modeling probability mass evolution and estimation performance, will be developed to efficiently represent and optimize multi-robot sensing control policies. Second, the project will demonstrate that a team of ground and aerial robots, using Active Bayesian Inference techniques, can achieve autonomous exploration and active high-fidelity mapping of an unknown environment. This objective will be supported by novel contributions to online dense implicit surface mapping in terms of distributed and probabilistic techniques that allow multiple robots to collaboratively estimate the environment geometry and semantics, while quantifying the uncertainty of these estimates to allow planning informative actions.This project is supported by the cross-directorate Foundational Research in Robotics program, jointly managed and funded by the Directorates for Engineering (ENG) and Computer and Information Science and Engineering (CISE).This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

人工感知技术允许机器人系统使用传感数据来了解它们的位置和周围环境，这对于在仔细控制的制造环境之外实现机器人自动化至关重要。然而，目前的机器人系统在感知世界方面仍然是被动的。与生物系统不同，机器人缺乏探索和缓解不确定性的好奇心机制，这对智能决策至关重要。这种能力在灾害应对、安全和监视以及环境监测方面非常重要，因为在这些方面，需要迅速了解环境中的地形、建筑物和人员的情况。在这个项目中开发的方法将影响自主机器人团队的映射和主动传感算法的设计及其在上述应用中的使用。这项教师早期职业发展（CAREER）计划的研究开发了基本的机器人自主能力，这也将影响依赖自主机器人的其他领域。此外，该项目将开发一套开源教育材料，包括理论问题，项目，讲座和核心机器人算法的示例性实现，统一在一个易于访问的模拟环境中。该平台将支持研究生的课程开发，以及本科生和K-12学生的外联和研究启动活动。研究议程将通过两项关键技术创新来实现。首先，该项目将正式定义一个主动贝叶斯推理问题，寻求最小不确定性估计的传感系统的最佳控制。分布式近似动态规划的方法，利用问题的结构，诱导的功能建模概率质量演化和估计性能，将开发有效地表示和优化多机器人传感控制策略。其次，该项目将证明，一组地面和空中机器人，使用主动贝叶斯推理技术，可以实现自主探索和主动高保真地图的未知环境。这一目标将得到在线密集隐式表面映射的分布式和概率技术的新贡献的支持，该技术允许多个机器人协作估计环境几何形状和语义，同时量化这些估计的不确定性，以允许规划信息行动。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Semantic OcTree Mapping and Shannon Mutual Information Computation for Robot Exploration

• DOI: 10.1109/tro.2023.3245986
• 发表时间: 2021-12
• 期刊: IEEE Transactions on Robotics
• 影响因子: 7.8
• 作者: Arash Asgharivaskasi;Nikolay A. Atanasov

Distributed Bayesian Estimation of Continuous Variables Over Time-Varying Directed Networks

时变有向网络连续变量的分布式贝叶斯估计

• DOI: 10.1109/lcsys.2022.3167654
• 发表时间: 2022
• 期刊: IEEE Control Systems Letters
• 影响因子: 3
• 作者: Paritosh, Parth;Atanasov, Nikolay;Martinez, Sonia
• 通讯作者: Martinez, Sonia

Active Mapping via Gradient Ascent Optimization of Shannon Mutual Information over Continuous SE(3) Trajectories

• DOI: 10.1109/iros47612.2022.9981875
• 发表时间: 2022-04
• 期刊: 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
• 作者: Arash Asgharivaskasi;Shumon Koga;Nikolay A. Atanasov

Learning Continuous Control Policies for Information-Theoretic Active Perception

学习信息论主动感知的连续控制策略

• DOI: 10.1109/icra48891.2023.10160455
• 发表时间: 2023
• 期刊: 2023 IEEE International Conference on Robotics and Automation (ICRA
• 作者: Yang, Pengzhi;Liu, Yuhan;Koga, Shumon;Asgharivaskasi, Arash;Atanasov, Nikolay
• 通讯作者: Atanasov, Nikolay

Policy Learning for Active Target Tracking over Continuous SE(3) Trajectories

连续 SE(3) 轨迹上主动目标跟踪的策略学习

• 发表时间: 2023
• 期刊: Learning for Dynamics and Control (L4DC
• 作者: Yang, Pengzhi;Koga, Shumon;Asgharivaskasi, Arash;Atanasov, Nikolay
• 通讯作者: Atanasov, Nikolay