Autonomous Social Robot Navigation using a Behavioral Finite State Social Machine

Robotica ◽  
2020 ◽  
Vol 38 (12) ◽  
pp. 2266-2289
Author(s):  
Vaibhav Malviya ◽  
Arun Kumar Reddy ◽  
Rahul Kala
Keyword(s):  
Navigation System ◽  
Robot Navigation ◽  
Field Of View ◽  
Artificial Potential Field ◽  
Social Response ◽  
Human Group ◽  
Local Planning ◽  
Time Duration ◽  
Finite State ◽  
Social Machine

SUMMARYWe present a robot navigation system based on Behavioral Finite State Social Machine. The paper makes a robot operate as a social tour guide that adapts its navigation based on the behavior of the visitors. The problem of a robot leading a human group with a limited field-of-view vision is relatively untouched in the literature. Uncertainties arise when the visitors are not visible, wherein the behavior of the robot is adapted as a social response. Artificial potential field is used for local planning, and a velocity manager sets the speed disproportional to time duration of missing visitors.

scholarly journals Towards Goal-Directed Navigation Through Combining Learning Based Global and Local Planners

Sensors ◽  
2019 ◽  
Vol 19 (1) ◽  
pp. 176 ◽  
Author(s):  
Xiaomao Zhou ◽  
Yanbin Gao ◽  
Lianwu Guan
Keyword(s):  
Reinforcement Learning ◽  
Navigation System ◽  
Fundamental Problem ◽  
Robot Navigation ◽  
Great Success ◽  
Navigation Systems ◽  
Local Planning ◽  
End To End ◽  
Global Planning ◽  
Global And Local

Robot navigation is a fundamental problem in robotics and various approaches have been developed to cope with this problem. Despite the great success of previous approaches, learning-based methods are receiving growing interest in the research community. They have shown great efficiency in solving navigation tasks and offer considerable promise to build intelligent navigation systems. This paper presents a goal-directed robot navigation system that integrates global planning based on goal-directed end-to-end learning and local planning based on reinforcement learning (RL). The proposed system aims to navigate the robot to desired goal positions while also being adaptive to changes in the environment. The global planner is trained to imitate an expert’s navigation between different positions by goal-directed end-to-end learning, where both the goal representations and local observations are incorporated to generate actions. However, it is trained in a supervised fashion and is weak in dealing with changes in the environment. To solve this problem, a local planner based on deep reinforcement learning (DRL) is designed. The local planner is first implemented in a simulator and then transferred to the real world. It works complementarily to deal with situations that have not been met during training the global planner and is able to generalize over different situations. The experimental results on a robot platform demonstrate the effectiveness of the proposed navigation system.

scholarly journals Multi-Instance Inertial Navigation System for Radar Terrain Imaging

2020 ◽  
Vol 12 (21) ◽  
pp. 3639
Author(s):  
Michal Labowski ◽  
Piotr Kaniewski
Keyword(s):  
Navigation System ◽  
Inertial Navigation System ◽  
Inertial Navigation ◽  
Navigation Systems ◽  
Time Duration ◽  
Navigation Data ◽  
Positioning Errors ◽  
Measurement Session ◽  
Single Aperture ◽  
Gps System

Navigation systems used for the motion correction (MOCO) of radar terrain images have several limitations, including the maximum duration of the measurement session, the time duration of the synthetic aperture, and only focusing on minimizing long-term positioning errors of the radar host. To overcome these limitations, a novel, multi-instance inertial navigation system (MINS) has been proposed by the authors. In this approach, the classic inertial navigation system (INS), which works from the beginning to the end of the measurement session, was replaced by short INS instances. The initialization of each INS instance is performed using an INS/GPS system and is triggered by exceeding the positioning error of the currently operating instance. According to this procedure, both INS instances operate simultaneously. The parallel work of the instances is performed until the image line can be calculated using navigation data originating only from the new instance. The described mechanism aims to perform instance switching in a manner that does not disturb the initial phases of echo signals processed in a single aperture. The obtained results indicate that the proposed method improves the imaging quality compared to the methods using the classic INS or the INS/GPS system.

Sign in / Sign up

Export Citation Format

Share Document