Please use this identifier to cite or link to this item: https://scholarbank.nus.edu.sg/handle/10635/151881
Title: PERFORMANCE EVALUATION AND COMPARISON OF COLLISION AVOIDANCE ALGORITHMS
Authors: YAZHINI CHITRA PRADEEP
ORCID iD:   orcid.org/0000-0001-6645-6674
Keywords: Performance metrics, Performance evaluation, Collision avoidance algorithms, Mobile robot navigation, Human-like behaviour, Human safety
Issue Date: 3-Aug-2018
Citation: YAZHINI CHITRA PRADEEP (2018-08-03). PERFORMANCE EVALUATION AND COMPARISON OF COLLISION AVOIDANCE ALGORITHMS. ScholarBank@NUS Repository.
Abstract: The objective of the thesis is to establish a standardized evaluation framework for quantifying the performance of collision avoidance algorithms with applications to autonomous vehicles and mobile robots. A novel performance metric referred to as Pc metric is proposed based on the concept of probability of collision by taking into account the stochastic and dynamic interactions between different objects involved in a mobile robotic environment. The usefulness of the metric is demonstrated in measuring the safety performance of robots and in inferring the safety felt by humans due to the robot's behaviour. Having realised the importance of human safety in a complex robotic environment, the thesis proposes a proxemics-based navigation algorithm, inspired from pedestrian studies, to mimic human-like responses in robot navigation. The developed Pc metric objectively demonstrated that the proposed navigation produces robot motion that is safe and intuitive to humans with the given minimal information of the environment.
URI: http://scholarbank.nus.edu.sg/handle/10635/151881
Appears in Collections:Ph.D Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
ChitraPradeepY.PDF14.32 MBAdobe PDF

OPEN

NoneView/Download

Page view(s)

44
checked on Jul 10, 2020

Download(s)

16
checked on Jul 10, 2020

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.