访问计数 309581 (自2016年5月)
Videos
0?1470885445
发布时间:2023-04-27 22:15
更新时间:2023-04-27 22:15
Experimental video for tour paper entitled "A Safe Reinforcement Learning Approach for Autonomous Navigation of Mobile Robots in Dynamic Environments"
( 31.848 MB) 周智千, 2023-04-27 21:43
回复 ︿
用户头像
登录后可添加回复
0?1470885445
发布时间:2022-06-28 02:09
更新时间:2022-07-21 22:00
In the fight against COVID-19, many robots replace human employees in various tasks that involve a risk of infection. Among these tasks, the fundamental problem of navigating robots among crowds, named robot crowd navigation, remains open and challenging. Therefore, we propose HGAT-DRL, a heterogeneous GAT-based deep reinforcement learning algorithm. This algorithm encodes the constrained human-robot-coexisting environment in a heterogeneous graph consisting of four types of nodes. It also constructs an interactive agent-level representation for objects surrounding the robot, and incorporates the kinodynamic constraints from the non-holonomic motion model into the deep reinforcement learning (DRL) framework. Simulation results show that our proposed algorithm achieves a success rate of 92%, at least 6% higher than four baseline algorithms. Furthermore, the hardware experiment on a Fetch robot demonstrates our algorithm's successful and convenient migration to real robots.
( 53.348 MB) 周智千, 2022-07-21 21:59
回复 ︿
用户头像
登录后可添加回复
0?1470885445
发布时间:2022-03-04 10:00
更新时间:2022-03-04 10:00

Crowd navigation has becoming an increasingly prominent problem in robotics. The main challenge comes from the lack of understanding of pedestrians’ behaviors. Encouraged by the great achievement in trajectory prediction, the twin field of crowd navigation, this work focus on integrating trajectory prediction with path planning and proposed a crowd navigation algorithm named RHC-T (Receding Horizon Control with Trajjectron++). It consists of two independent modules: one for trajectory prediction and another for receding horizon control. Benefiting from the trajectory prediction module, RHC-T builds up an explicit understanding of pedestrians’behaviors in the form of predicted trajectories. Base on the formulation of receding horizon control, the proposed algorithm can deal with the time-varying obstacle constraints from pedestrians, naturally. Furthermore, extensive experiments are performed on two pedestrian trajectory datasets, ETH and UCY, to evaluate the proposed algorithm in a more realistic way than previous works. Experimental results show that RHC-T reduces the intervention to pedestrians significantly and navigates the robot in time-efficient paths. Compared with three baseline algorithms, RHC-T achieves better performance with an improvement in the intervention rate and navigation time of at least 8.00% and 3.88%, respectively.


image


回复 ︿
用户头像
登录后可添加回复
0?1470885445
发布时间:2022-03-01 14:09
更新时间:2022-03-01 14:18

This video is the accompanying video of the paper: Jiayang Liu, Junhao Xiao, Huimin Lu, Zhiqian Zhou, Sichao Lin, Zhiqiang Zheng. Terrain Assessment Based on Dynamic Voxel Grids in Outdoor Unstructured Environments


Abstract: For ground robots working in outdoor unstructured environments, terrain assessment is a key step for path planning.In this paper, we propose a novel terrain assessment method. The raw 3D point clouds are segmented based on dynamic voxel grids, then the untraversable areas are extracted and stored in the form of 2D occupancy grid maps. Afterwards, only the traversable areas are processed and stored in the form of 2.5D digital elevation maps (DEMs). In this case, the efficiency of the terrain assessment is improved and the query space of terrain feature information is reduced. To evaluate the proposed algorithm, the approach operating on point clouds has served as the baseline. According to the experimental results, our method has a better performance in both assessment time and query efficiency.



回复 ︿
用户头像
登录后可添加回复
0?1470885445
李筱 TO  NuBot Research Team | Videos
发布时间:2019-04-25 09:20
更新时间:2019-04-25 17:25

This video is the accompanying video of the paper:Xiao Li, Bingxin Han, Zhiwen Zeng, Junhao Xiao, Huimin Lu. Human-Robot Interaction Based on Battle Management Language for Multi-robot System


Abstract: Commanding and controlling a multi-robot system is a challenging task. Static control commands are difficult to fully meet the requirements of controlling different robots. As the number of robots increases, it is difficult for the robot's motion-level commands to simultaneously satisfy the demands of commanding multi-robot system. This paper uses a limited natural language to control multi-robot systems, and proposes a framework based on Battle Management Language (BML) to command multi-robot systems. Based on the framework, the capabilities and names of the robot can be dynamically added to the dictionary, and the limited natural language can be converted into a standard BML command according to the dictionary to control the multi-robot system. In this way, the robot can execute motion-level commands, such as movement, steering, etc., and can also perform task-level commands, such as enclosing, defense, etc. The experimental results show that the system composed of different types of robots can be commanded by using the interactive framework proposed in this paper.


( 137.992 MB) 李筱, 2019-04-25 09:18
回复 ︿
用户头像
登录后可添加回复
0?1470885445
李筱 TO  NuBot Research Team | Videos
发布时间:2019-01-27 19:56
更新时间:2019-01-27 19:56
qualification video 2019
( 91.018 MB) 李筱, 2019-01-27 19:55
回复 ︿
用户头像
登录后可添加回复
0?1470885445
李义 TO  NuBot Research Team | Videos
发布时间:2018-07-10 09:46
更新时间:2018-11-22 00:39

This video is about the experimental results of the following paper: Yi Li, Chenggang Xie, Huimin Lu, Xieyuanli Chen, Junhao Xiao and Hui Zhang. Scale-aware Monocular SLAM Based on Convolutional Neural Network. Proceedings of the 15th IEEE International Conference on Information and Automation 2018 ( ICIA 2018 ), Mount Wuyi, 2018.


Abstract—Remarkable performance has been achieved using the state-of-the-art monocular Simultaneous Localization and Mapping (SLAM) algorithms. However, due to the scale ambiguity limitation of monocular vision, the existing monocular SLAM systems can not directly restore the absolute scale in unknown environments. Given the amazing results in the field of depth estimation from Convolutional Neural Networks (CNNs), we propose a CNN-based monocular SLAM, where we naturally combine the CNN-predicted depth maps together with the monocular ORB-SLAM, overcoming the scale ambiguity limitation of the monocular SLAM. We test our method using the popular KITTI odometry benchmark, and the experimental results show that the overall performance of average translational and rotational error can reach 2.00% and 0.0051º/m. In addition, our approach can work well under the pure rotation motion, which shows the robustness and high accuracy of the proposed algorithm.

回复 ︿
用户头像
登录后可添加回复
0?1470885445
发布时间:2018-04-08 10:59
更新时间:2018-04-08 15:02

Abstract— Most robots in urban search and rescue (USAR) fulfill tasks teleoperated by human operators. The operator has to know the location of the robot and find the position of the target (victim). This paper presents an augmented reality system using a Kinect sensor on a customly designed rescue robot. Firstly, Simultaneous Localization and Mapping (SLAM) using RGB-D cameras is running to get the position and posture of the robot. Secondly, a deep learning method is adopted to obtain the location of the target. Finally, we place an AR marker of the target in the global coordinate and display it on the operator's screen to indicate the target even when the target is out of the camera’s field of view. The experimental results show that the proposed system can be applied to help humans interact with robots.

回复 ︿
用户头像
登录后可添加回复
0?1470885445
Junchong TO  NuBot Research Team | Videos
发布时间:2018-01-30 13:40
更新时间:2018-04-08 10:47

This video is the accompanying video of the paper: Junchong Ma, Weijia Yao, Wei Dai, Huimin Lu, Junhao Xiao, Zhiqiang Zheng. Cooperative Encirclement Control for a Group of Targets by Decentralized Robots with Collision Avoidance. Proceedings of the 37th Chinese Control Conference, 2018.

Abstract: This study focuses on multi-target capture and encirclement control problem for multiple mobile robots. With the distributed architecture, this problem involves a group of robots to encircle several moving targets in a coordinated circle formation. In order to efficiently allocate the targets to robots, a Hybrid Dynamic Task Allocation (HDTA) algorithm was proposed, in which a temporary "manager" robot was assigned to negotiate with other robots. For encirclement formation, a robust  control law was introduced for any number of mobile robots to form a specific circle formation with arbitrary inter-robot angular spacing. In view of safety, an online collision avoidance algorithm combining the sub-targets and Artificial Potential Fields (APF) approaches was proposed, which ensures that the paths of robots are collision-free. To prove the validity and robustness of the proposed scheme, both theoretical analysis and simulation experiments were conducted.

回复 ︿
用户头像
登录后可添加回复
0?1470885445
发布时间:2016-06-23 22:35
更新时间:2017-12-22 17:37

This video is the accompanying video of the paper: Yi Liu, Yuhua Zhong, Xieyuanli Chen, Pan Wan, Huimin Lu, Junhao Xiao, Hui Zhang, The Design of a Fully Autonomous Robot System for Urban Search and Rescue, Proceedings of the 2016 IEEE International Conference on Information and Automation, 2016.

Abstract: Autonomous robots in urban search and rescue (USAR) have to fulfill several tasks at the same time: localization, mapping, exploration, object recognition, etc. This paper describes the whole system and the underlying research of the NuBot rescue robot for participating RoboCup Rescue competition, especially in exploring the rescue environment autonomously. A novel path following strategy and a multi-sensor based controller are designed to control the robot for traversing the unstructured terrain. The robot system has been successfully applied and tested in the RoboCup Rescue Robot League (RRL) competition and won the championship of 2016 RoboCup China Open RRL competition.


回复 ︿
用户头像
登录后可添加回复
点击展开更多