2008 IEEE International Conference on Technologies for Practical Robot Applications 2008
DOI: 10.1109/tepra.2008.4686678
|View full text |Cite
|
Sign up to set email alerts
|

Centralized processing and distributed I/O for robot control

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
6
0

Year Published

2008
2008
2018
2018

Publication Types

Select...
6
1

Relationship

2
5

Authors

Journals

citations
Cited by 14 publications
(6 citation statements)
references
References 5 publications
0
6
0
Order By: Relevance
“…Given the bus speed of 400 Mbps, we concluded that software overheads were a predominant factor in the latency, as in [13]. As we concluded in [19], it became necessary to bundle the data for multiple axes into blocks in order to overcome these limitations. Figure 7 shows a raw sampling of read/write times over the full range of block sizes (in quadlets, the smallest unit in IEEE 1394), up to the maximum of 512 quadlets for the 400 Mbps mode.…”
Section: Resultsmentioning
confidence: 80%
See 1 more Smart Citation
“…Given the bus speed of 400 Mbps, we concluded that software overheads were a predominant factor in the latency, as in [13]. As we concluded in [19], it became necessary to bundle the data for multiple axes into blocks in order to overcome these limitations. Figure 7 shows a raw sampling of read/write times over the full range of block sizes (in quadlets, the smallest unit in IEEE 1394), up to the maximum of 512 quadlets for the 400 Mbps mode.…”
Section: Resultsmentioning
confidence: 80%
“…The high speed serial bus encourages the distributed I/O and centralized processing architecture we presented in [19]. One advantage of this approach is that the I/O processing logic is simple and requires little maintenance.…”
Section: Motivations and Objectivesmentioning
confidence: 99%
“…In [6], a distributed I/O with centralized processing architecture using a communication network based on IEEE 1394 (FireWire) is presented. This kind of system is very suitable for research purposes as it keeps the peripheral management simple, while maintaining a reduced number of cables.…”
Section: Related Workmentioning
confidence: 99%
“…Finally, despite the high-speed transmission and reduced cabling of FireWire, as presented in [6], it has a reduced performance when implementing several nodes due to the high fixed latency in read and write operations.…”
Section: B Network Systemmentioning
confidence: 99%
“…Most existing controllers use multi-processing; one compelling reason is that computation is often physically distributed. We advocate a new approach, where processing is centralized and I/O is distributed [1]. We contend that this is feasible due to advances in processor performance, especially multicore architectures, coupled with the high data rates of modern serial networks.…”
Section: Introductionmentioning
confidence: 99%