Abstract
Autonomous robot navigation and dynamic obstacle avoidance in complex, cluttered, indoor environments is a challenging task. A robust solution would allow robots to be deployed in hospitals, airports or shopping centres to serve as guides and fulfil other functions requiring safe human--robot interaction. Previous studies have explored various approaches to selecting sensor types, collecting data, and training models capable of safely avoiding unmapped, possibly dynamic obstacles in an indoor environment. In this paper we address the problem of recognizing and anticipating collisions, in order to determine when avoidance manoeuvres are required. We propose and compare two sensor-fusion and neural-network-based solutions, one in which models are trained separately on static and dynamic samples and another in which a model is trained on samples of collisions with both dynamic and static obstacles. The measured accuracies confirmed that the separately trained, ensemble models had better recognition performance, but were slower at calculation than the models trained without taking the obstacle types into account.
Originalsprog | Engelsk |
---|---|
Titel | ESANN 2019 - Proceedings, 27th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning |
Redaktører | Michel Verleysen |
Antal sider | 6 |
Udgivelsessted | Belgium |
Forlag | ESANN |
Publikationsdato | 28 apr. 2019 |
Sider | 655-660 |
Kapitel | Dynamical systems and reinforcement learning |
ISBN (Trykt) | 9782875870650 |
ISBN (Elektronisk) | 9782875870667 |
Status | Udgivet - 28 apr. 2019 |
Begivenhed | European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning - Bruges, Belgien Varighed: 24 apr. 2019 → 26 apr. 2019 Konferencens nummer: 27 https://www.elen.ucl.ac.be/esann/ |
Konference
Konference | European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning |
---|---|
Nummer | 27 |
Land/Område | Belgien |
By | Bruges |
Periode | 24/04/2019 → 26/04/2019 |
Internetadresse |