2004
DOI: 10.1109/jproc.2003.823157
|View full text |Cite
|
Sign up to set email alerts
|

Abstract: This paper shows how state-of-the-art state estimation techniques can be used to provide efficient solutions to the difficult problem of real-time diagnosis in mobile robots. The power of the adopted estimation techniques resides in our ability to combine particle filters with classical algorithms, such as Kalman filters. We demonstrate these techniques in two scenarios: a mobile waiter robot and planetary rovers designed by NASA for Mars exploration.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
54
0

Year Published

2011
2011
2019
2019

Publication Types

Select...
4
3

Relationship

0
7

Authors

Journals

citations
Cited by 75 publications
(54 citation statements)
references
References 23 publications
0
54
0
Order By: Relevance
“…One way to overcome this is to use particle filters. 14,15 Although model-based approaches produce powerful tools to detect and identify faults, their design relies heavily on expert knowledge and therefore requires significant resources to develop and implement onboard complex systems.…”
Section: Model-basedmentioning
confidence: 99%
See 1 more Smart Citation
“…One way to overcome this is to use particle filters. 14,15 Although model-based approaches produce powerful tools to detect and identify faults, their design relies heavily on expert knowledge and therefore requires significant resources to develop and implement onboard complex systems.…”
Section: Model-basedmentioning
confidence: 99%
“…We compute the similarity between each temporal neighborhood and the topics learned from the training datasets using KL, and we classify the temporal neighborhoods using the semantic labels associated with the most similar topic (i.e., the nearest-neighbor; Section 3.6) in a dataset. We have chosen to use this method over other algorithms that can infer the number of clusters from the data, such as spectral clustering 22 and affinity propagation, 14 mainly because it is a fully Bayesian generative probabilistic model. As such, the method offers an inherent uncertainty criterion for estimating the clustering quality of the model and its ability to generalize to unseen data.…”
Section: Data-drivenmentioning
confidence: 99%
“…The look-ahead strategy altogether with a technique of marginalising out some of the variables is then collectively called the look-ahead Rao-Blackwellised particle filter (la-RBPF) and is applied for fault diagnostic tasks in two industrial processes: an industrial dryer and a level-tank [3]. Also in [5], the la-RBPF is adopted for detecting faults in the autonomous operation of a mobile waiter robot and planetary rovers designed by NASA for Mars exploration. The results therein show how the la-RBPF outperforms existing particle filters in decreasing of diagnosis errors and variances especially when the number of particles is relatively low, i.e., < 100.…”
Section: Introductionmentioning
confidence: 99%
“…This is mainly due to perturbation in the form of likelihood of all possible discrete states generated in the process of UKF updates. Even though in fault diagnosis problems whose number of discrete states are usually low, la-URBPF will fail miserably in high noise scenarios and the expected benefits of using la-URBPF instead of its plain URBPF counterparts is no further relevant [5], [9]. Simply increasing the number of particles (e.g., from 10 to 50) cannot improve the situation in such noisy environments [9].…”
Section: Introductionmentioning
confidence: 99%
“…In RBPF algorithm, the state space is divided into linear substate space estimated by Kalman filter algorithm and non-linear sub-state space estimated by particle filter algorithm [15][16][17][18]. The RBPF has applications in the hybrid Gaussian [19][20][21], fixed parameters estimation [22], hidden Markov models (HMMs) [20][21], Dirichlet process models and Dynamic Bayesian Networks (DBN) [23]. There is no transition function for the unknown parameter in the state space equations.…”
Section: Introductionmentioning
confidence: 99%