Object Identification using Support Vector Regression for Haptic Object Reconstruction

Dewapura, Praveena Wimarshani; Jayawardhana, Kemantha; Harsha, A. M.; Abeykoon, A. M. Harsha S.

doi:10.1109/eecon52960.2021.9580959

Cited by 3 publications

(3 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Despite the drawbacks [8], [9], strain gauges were used to measure the force. A Support Vector Regression (SVR) model to identify the object to be reconstructed was introduced and it has shown higher performance than the traditional model-based approach [14]. However, this model has considered only position and velocity information to predict force response.…”

Section: Introductionmentioning

confidence: 99%

Haptic feature selection using Statistical Analysis for Haptic Object Reconstruction

Dewapura

Abeykoon

2023

Preprint

View full text Add to dashboard Cite

Object reconstruction in the haptic dimension is crucial in achieving realization in the virtual world. Prior studies have employed the conventional spring damper model to interpret the behavior of the object and motion parameters were treated as determinants of force feedback. However, haptic information is nonlinear, and more factors could influence haptic feedback recreation than just motion parameters. In recent years, scientists have paid much attention on employing AI technologies with haptics. AI approach is data-driven, and a thorough understanding of the data is critical to its performance. As a result, a detailed statistical analysisis required to discover patterns and relationships within the features and identify significant features as utilizing all the factors affecting haptic object reconstruction will cost higher computational time and resources. Thus, this study focused on statistically analyzing data extracted through a Disturbance Observer (DOB) and Reaction Force Observer(RFOB) based sensorless technique and selecting the significant features to precisely reconstruct an object in a haptic dimension. Furthermore, a mathematical notation was introduced to define a feature and the feature matrix was derived using the notation. Finally, a comparison of the performance of the same AI algorithm was conducted by varying the feature input to the AI model to understand the effect of the feature matrix.

show abstract

Section: Introductionmentioning

confidence: 99%

Haptic feature selection using Statistical Analysis for Haptic Object Reconstruction

Dewapura

Abeykoon

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…However, they have relied on strain gauges to measure force [8], [9]. A Support Vector Regression (SVR) model was introduced to identify objects for reconstruction and it has shown higher performance than the traditional model-based approach though it only considered position and velocity information to predict force response [14].…”

Section: Introductionmentioning

confidence: 99%

Haptic feature selection using Statistical Analysis for Haptic Object Reconstruction

Dewapura

Abeykoon

2023

Preprint

Self Cite

View full text Add to dashboard Cite

Object reconstruction in the haptic dimension is crucial in achieving realization in the virtual world. Prior studies have employed the conventional spring damper model to interpret object behavior, with motion parameters being the main deter- minant of force feedback. However, haptic information is nonlinear and more factors could influence haptic feedback recreation than just motion parameters. In recent years, scientists have paid much attention to employing AI technologies with haptics. AI approach is data-driven, and a thorough understanding of the data is critical to its performance. As a result, a detailed statistical analysis is required to discover patterns and relationships within the features and identify significant features as utilizing all the factors affecting haptic object reconstruc- tion will cost higher computational time and resources. Thus, this study focused on statistically analyzing data extracted through a Disturbance Observer (DOB) and Reaction Force Observer (RFOB) based sensorless technique and selecting the significant features to reconstruct an object in a haptic dimension precisely. Furthermore, a mathematical notation was introduced to define a feature and the feature matrix was derived using the notation. Finally, the performance of an AI algorithm was then compared using different feature inputs to the AI model to understand the effect of the feature matrix.

show abstract

“…Notably, the nonlinear structures of these models are rigid to some extent, thus requiring prior knowledge about the system structure. Furthermore, black-box models, for instance, wavelet [24,25], neural networks [26][27][28][29][30], and support vector machines [31][32][33], have also been employed, which lack interpretability and suffer from the curse of dimensionality [34]. Furthermore, due to the lack of transparency of the models, they are not very useful for model-based control system design.…”

Section: Introductionmentioning

confidence: 99%

Enhancement of the HILOMOT Algorithm with Modified EM and Modified PSO Algorithms for Nonlinear Systems Identification

Mahfuz

Muyeen

2022

Electronics

View full text Add to dashboard Cite

Developing a mathematical model has become an inevitable need in studies of all disciplines. With advancements in technology, there is an emerging need to develop complex mathematical models. System identification is a popular way of constructing mathematical models of highly complex processes when an analytical model is not feasible. One of the many model architectures of system identification is to utilize a Local Model Network (LMN). Hierarchical Local Model Tree (HILOMOT) is an iterative LMN training algorithm that uses the axis-oblique split method to divide the input space hierarchically. The split positions of the local models directly influence the accuracy of the entire model. However, finding the best split positions of the local models presents a nonlinear optimization problem. This paper presents an optimized HILOMOT algorithm with enhanced Expectation–Maximization (EM) and Particle Swarm Optimization (PSO) algorithms which includes the normalization parameter and utilizes the reduced-parameter vector. Finally, the performance of the improved HILOMOT algorithm is compared with the existing algorithm by modeling the NOx emission model of a gas turbine and multiple nonlinear test functions of different orders and structures.

show abstract

Object Identification using Support Vector Regression for Haptic Object Reconstruction

Cited by 3 publications

References 16 publications

Haptic feature selection using Statistical Analysis for Haptic Object Reconstruction

Haptic feature selection using Statistical Analysis for Haptic Object Reconstruction

Haptic feature selection using Statistical Analysis for Haptic Object Reconstruction

Enhancement of the HILOMOT Algorithm with Modified EM and Modified PSO Algorithms for Nonlinear Systems Identification

Contact Info

Product

Resources

About