Human-Aware Robot Navigation by Behavioral Metrics Based on Predictive Model

Ranasinghe, Anuradha and Dasgupta, Prokar and Nanayakkara, Thrishantha (2018) Human-Aware Robot Navigation by Behavioral Metrics Based on Predictive Model. In: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2018), 1st to 5th October 2018, Madrid Spain.

[img] Text
Human-Aware Robot Navigation by Behavioral Metrics Based on Predictive Model.pdf - Accepted Version

Download (406kB)
Official URL: https://www.iros2018.org/

Abstract

Human-aware robot navigation is very important in many applications in human-robot shared environments. There are some situations, people have to move with less visual and auditory perceptions. In that case, the robot can help to enhance the efficiency of navigation when moving in noisy and low visibility conditions. In that scenario, haptic is the best way to communicate when other modalities are less reliable. We used a rein to guide a human when 1-DoF robotic arm can perturb the humans’ arm to guide into a desired point. The novelty of our work is presenting behavioral metrics based on novel predictive model to strategically position the humans in human-robot shared environment in low visibly and auditory conditions. We found that humans start with a second order reactive autoregressive following model and changes it to a predictive model with training. This result would help us to enhance humans’ safety and comfort in robot leading navigation in shared environment.

Item Type: Conference or Workshop Item (Speech)
Faculty / Department: Faculty of Science > Mathematics and Computer Science
Depositing User: Anuradha Dissanayake Mudiyanselage
Date Deposited: 02 Jun 2019 14:49
Last Modified: 02 Jun 2019 14:49
URI: http://hira.hope.ac.uk/id/eprint/2870

Actions (login required)

View Item View Item