12,756 Hits in 6.7 sec

Sample-Efficient Training of Robotic Guide Using Human Path Prediction Network [article]

Hee-Seung Moon, Jiwon Seo
2020 arXiv   pre-print
This paper proposes a human path prediction network (HPPN) and an evolution strategy-based robot training method using virtual human movements generated by the HPPN, which compensates for this sample inefficiency  ...  This sample-efficient training method is expected to be widely applicable to all robots and computing machinery that physically interact with humans.  ...  We first train a human path prediction network with a limited number of episodes, and then train the robot policy using numerous virtual episodes generated based on the human path prediction network.  ... 
arXiv:2008.05054v1 fatcat:skzlgyjo7za2pakbqhs7zmqqnm

Prediction of Human Trajectory Following a Haptic Robotic Guide Using Recurrent Neural Networks [article]

Hee-Seung Moon, Jiwon Seo
2019 arXiv   pre-print
In this paper, we present a method for predicting the trajectory of a human who follows a haptic robotic guide without using sight, which is valuable for assistive robots that aid the visually impaired  ...  We apply a deep learning method based on recurrent neural networks using multimodal data: (1) human trajectory, (2) movement of the robotic guide, (3) haptic input data measured from the physical interaction  ...  Prediction of Human Trajectory Following a Haptic Robotic Guide Using Recurrent Neural Networks Hee-Seung Moon and Jiwon Seo, Member, IEEE To measure accurate haptic input from the user, our robotic-guide  ... 
arXiv:1903.01027v1 fatcat:d3vnoihad5aw3prvlewrvm4l34

Prediction of Bottleneck Points for Manipulation Planning in Cluttered Environment using a 3D Convolutional Neural Network [article]

Indraneel Patil, B.K. Rout, V. Kalaichelvi
2019 arXiv   pre-print
Latest research in industrial robotics is aimed at making human robot collaboration possible seamlessly.  ...  relevant context extracted from the planning scenario using a 3D Convolutional Neural Network and it is also proven that the proposed technique generalises to unseen problem instances.  ...  For testing and training datasets during transfer learning a 1:9 ratio was used. B. Network Architecture and Training The 3D CNN architecture is a modified version of the VoxNet network.  ... 
arXiv:1911.04676v1 fatcat:cm2oy2nthbb2veisnthyrdzvna

Learning World Transition Model for Socially Aware Robot Navigation [article]

Yuxiang Cui, Haodong Zhang, Yue Wang, Rong Xiong
2020 arXiv   pre-print
The navigation policy is trained with both real interaction data from multi-agent simulation and virtual data from a deep transition model that predicts the evolution of surrounding dynamics of mobile  ...  Experiments were conducted in multiple social scenarios both in simulation and on real robots, the learned policy can guide the robots to the final targets successfully while avoiding pedestrians in a  ...  Traditional solution to this problem is a modular pipeline that consists of human detection, trajectory prediction and local path planning.  ... 
arXiv:2011.03922v1 fatcat:ki6buukqdfa6hjwtuot42d3i4i

EWareNet: Emotion Aware Human Intent Prediction and Adaptive Spatial Profile Fusion for Social Robot Navigation [article]

Venkatraman Narayanan and Bala Murali Manoghar and Rama Prashanth RV and Aniket Bera
2020 arXiv   pre-print
Our approach predicts the trajectory-based pedestrian intent from historical gaits, which is then used for intent-guided navigation taking into account social and proxemic constraints.  ...  We outperform current state-of-art algorithms for intent prediction from 3D gaits.  ...  The navigation model of [3, 25] uses the predicted emotions and use a constant multiplier to maintain proper comfort space with the humans while planning the robot path.  ... 
arXiv:2011.09438v3 fatcat:b3x2pdzlq5er5gqpxhwps4ddea

Segmenting areas of potential contamination for adaptive robotic disinfection in built environments

Da Hu, Hai Zhong, Shuai Li, Jindong Tan, Qiang He
2020 Building and Environment  
Third, with short-wavelength ultraviolet light, the trajectories of robotic disinfection are generated to adapt to the geometries of areas of potential contamination to ensure complete and safe disinfection  ...  Both simulations and physical experiments were conducted to validate the proposed methods, which demonstrated the feasibility of intelligent robotic disinfection and highlighted the applicability in mass-gathering  ...  The results demonstrate the efficiency and effectiveness of the robot path planning method. Fig. 15 presents two representative examples of the path planning of the robot.  ... 
doi:10.1016/j.buildenv.2020.107226 pmid:32868961 pmcid:PMC7448966 fatcat:kna5paa4cbgdviam75sxlwszli

ProxEmo: Gait-based Emotion Learning and Multi-view Proxemic Fusion for Socially-Aware Robot Navigation [article]

Venkatraman Narayanan, Bala Murali Manoghar, Vishnu Sashank Dorbala, Dinesh Manocha, Aniket Bera
2020 arXiv   pre-print
Our approach predicts the perceived emotions of a pedestrian from walking gaits, which is then used for emotion-guided navigation taking into account social and proxemic constraints.  ...  We highlight its benefits in terms of navigation in indoor scenes using a Clearpath Jackal robot.  ...  Social Robotics and Emotionally-Guided Navigation As robots have become more commonplace, their impact on humans' social lives has emerged as an active area of research.  ... 
arXiv:2003.01062v2 fatcat:6gro733cibdqrg65a37wggwbwi

Informative Path Planning for Active Learning in Aerial Semantic Mapping [article]

Julius Rückin, Liren Jin, Federico Magistri, Cyrill Stachniss, Marija Popović
2022 arXiv   pre-print
This enables us to adaptively guide a UAV to gather the most informative terrain images to be labelled by a human for model training.  ...  Our experimental evaluation on real-world data shows the benefit of using our informative planning approach in comparison to static coverage paths in terms of maximising model performance and reducing  ...  Informative Path Planning We develop informative path planning algorithms to guide a UAV to adaptively collect useful training data for our FCN.  ... 
arXiv:2203.01652v1 fatcat:7cfx3v4dazhydbaz2rhcx62eqm

Deep compositional robotic planners that follow natural language commands [article]

Yen-Ling Kuo, Boris Katz, Andrei Barbu
2020 arXiv   pre-print
We demonstrate how a sampling-based robotic planner can be augmented to learn to understand a sequence of natural language commands in a continuous configuration space to move and manipulate objects.  ...  A recurrent hierarchical deep network controls how the planner explores the environment, determines when a planned path is likely to achieve a goal, and estimates the confidence of each move to trade off  ...  Given a training set of commands and paths, one component network is trained per word in the command that the robot is following.  ... 
arXiv:2002.05201v2 fatcat:5ijjhh353vdl7et3oz5hguayda

Learning-based Fast Path Planning in Complex Environments [article]

Jianbang Liu, Baopu Li, Tingguang Li, Wenzheng Chi, Jiankun Wang, Max Q.-H. Meng
2021 arXiv   pre-print
However, our proposed framework can overcome this difficulty by using a learning-based prediction module and a sampling-based path planning module.  ...  Incorporated with a sampling-based path planner, we can extract a feasible path from the output image so that the robot can track it from start to goal.  ...  The generated promising region is used to guide the sampling process of the path planner, resulting in an efficient search of the state space.  ... 
arXiv:2110.10041v1 fatcat:bl4jllchqzbtvdzc5ubr5onpgy

Efficient Object Manipulation to an Arbitrary Goal Pose: Learning-based Anytime Prioritized Planning [article]

Kechun Xu, Hongxiang Yu, Renlang Huang, Dashun Guo, Yue Wang, Rong Xiong
2022 arXiv   pre-print
We learn an offline-training path cost estimator to predict approximate path planning costs, which serve as pseudo rewards to allow for pre-training the high-level planner without interacting with the  ...  We focus on the task of object manipulation to an arbitrary goal pose, in which a robot is supposed to pick an assigned object to place at the goal position with a specific orientation.  ...  Finishing this task seems easy and efficient for human, but raises a challenge on the complexity of robotic planning.  ... 
arXiv:2109.10583v2 fatcat:vpsk5nb2hjfrpo47xak7yukiq4

Autonomous Navigation Using Deep Reinforcement Learning in ROS

Ganesh Khekare, Shahrukh Sheikh
2021 International Journal of Artificial Intelligence and Machine Learning  
This problem is challenging with the highly stochastic behavior of people. Previous methods believe to provide features of human behavior, but these features vary from person to person.  ...  For an autonomous robot to move safely in an environment where people are around and moving dynamically without knowing their goal position, it is required to set navigation rules and human behaviors.  ...  Now, it takes the portion of state-value pair sampled from E, Eb. And performs an update on value network using the back-propagation technique.  ... 
doi:10.4018/ijaiml.20210701.oa4 fatcat:pqakfbfctzekxpmo323kpmkdhy

Vision-Based Robot Path Planning with Deep Learning [chapter]

Ping Wu, Yang Cao, Yuqing He, Decai Li
2017 Lecture Notes in Computer Science  
In this paper, a new method based on deep convolutional neural network (CNN) for path planning of robot is proposed, the aim of which is to transform the mission of path planning into a task of environment  ...  Finally, according to the results of classification, the moving direction of robots is exported.  ...  This work was supported by the Nature Science Foundation of China (Grant Nos. U1608253 and 61473282) and by the Liaoning Provincial Social Planning Found (L15BGL017).  ... 
doi:10.1007/978-3-319-68345-4_9 fatcat:fcoayqupqzfwdo4ofsoqqy57nq

Information-Driven Path Planning

Shi Bai, Tixiao Shan, Fanfei Chen, Lantao Liu, Brendan Englot
2021 Current Robotics Reports  
learning can be used to improve the robustness and efficiency of informative path planning in robotics.  ...  Purpose of Review The era of robotics-based environmental monitoring has given rise to many interesting areas of research.  ...  used to guide a map-building robot to uncover unknown regions.  ... 
doi:10.1007/s43154-021-00045-6 fatcat:cfnlvoaacjhptlbdtk6fhjm37m

A Virtual End-To-End Learning System for Robot Navigation based on Temporal Dependencies

Yanqiu Zhang, Ruiquan Ge, Lei Lyu, Jinling Zhang, Chen Lyu, Xiaojuan Yang
2020 IEEE Access  
Xu et al. used the FCN-LSTM model to process large-scale video data for path prediction [11] .  ...  Chowdhuri S et al. used multimodal multitask learning to learn multiple different behavioral modes in a single depth neural network [19] to train different neural networks to predict the steering angle  ... 
doi:10.1109/access.2020.3010695 fatcat:g5zbwi356zcl5eo5smfzrgpwbq
« Previous Showing results 1 — 15 out of 12,756 results