1990
DOI: 10.1007/bf00368972
|View full text |Cite
|
Sign up to set email alerts
|

Neural networks in robotics: A survey

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
17
0

Year Published

1991
1991
2012
2012

Publication Types

Select...
4
3
2

Relationship

0
9

Authors

Journals

citations
Cited by 51 publications
(17 citation statements)
references
References 33 publications
0
17
0
Order By: Relevance
“…The PC gets capture image and get the ( , , , ) x y z δ with help of image processing technology, the position of forearm end actuator is at 0 P (which is equal to 6 P of last trajectory planning). For the band carrier is keep on moving at a constant speed, the PC estimated time and position 2 P of end actuator could arrival to pick the workpiece. Then the end actuator move to 1 P which is exactly above 2 P , and down to 2 P , pick the workpiece, move to 3 P , move to 4 P which is right above destination point 5 P , then down to 5 P and release the workpiece, and move the end actuator to 6 P .…”
Section: Scara Robot Trajectory Definementioning
confidence: 99%
See 1 more Smart Citation
“…The PC gets capture image and get the ( , , , ) x y z δ with help of image processing technology, the position of forearm end actuator is at 0 P (which is equal to 6 P of last trajectory planning). For the band carrier is keep on moving at a constant speed, the PC estimated time and position 2 P of end actuator could arrival to pick the workpiece. Then the end actuator move to 1 P which is exactly above 2 P , and down to 2 P , pick the workpiece, move to 3 P , move to 4 P which is right above destination point 5 P , then down to 5 P and release the workpiece, and move the end actuator to 6 P .…”
Section: Scara Robot Trajectory Definementioning
confidence: 99%
“…Trajectory planning is the process of defining a desired path in joint or Cartesian space for joints or gripper, respectively [2]. Trajectory planning quality decided the robot system operation performance and efficiency.…”
Section: Trajectory Planning For Control Systemmentioning
confidence: 99%
“…The bottleneck of sequential processing inherent to conventional computer architectures can be overcome by the use of parallel processing techniques, which allow a significant increase in performance in both conventional [29] and neural net approaches to robot control [24]. Conventional approaches using parallel processing have to solve the task assignment problem, which can be achieved only by detailed source level analysis of algorithmic structure and task dependencies (291.…”
Section: Real-time Control Using Adaptive Load Distributionmentioning
confidence: 99%
“…Since the number of architectural options for ANNs is overwhelming the need arises to narrow down the options to representative neural architectures from these two categories. Consequently three well-known network architectures namely FF [16], EL network [17] and NARX [18] networks are selected that represent non-recurrent, i.e. feedforward, partially recurrent and fully recurrent networks, respectively.…”
Section: Introductionmentioning
confidence: 99%