Sensor Fusion for Distance Estimation Under Disturbance with Reflective Optical Sensors using Multi Layer Perceptron (MLP)

Mesa, Joao; Vasquez, Daniel; Valencia-Aguirre, Juliana; Valencia, Juan Sebastián Botero

doi:10.1109/tla.2019.8931134

Cited by 5 publications

(6 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(MLP) and Reflective Optical Sensors [89] Low cost, high accuracy and anti-interference ability, flexible architecture choice, suitable for embedded systems.…”

Section: Distance Estimation Methods Combining Multilayer Perceptronmentioning

confidence: 99%

“…However, the automatic gain control mechanism introduces errors, and localization accuracy is susceptible to varying environmental conditions [88]. Mesa et al [89] developed a distance estimation approach that harnesses the power of MultiLayer Perceptrons (MLPs) combined with a trio of reflective optical distance sensors-visible light, ultraviolet, and near infrared. This sensor fusion model is designed to extend the measurement capabilities and ensure redundancy, thereby improving accuracy and reducing susceptibility to interference by compensating for different radiation effects.…”

Section: Infrared Sensorsmentioning

confidence: 99%

See 1 more Smart Citation

Robotics Perception and Control: Key Technologies and Applications

Luo,

Zhou,

Zeng

et al. 2024

Micromachines

View full text Add to dashboard Cite

The integration of advanced sensor technologies has significantly propelled the dynamic development of robotics, thus inaugurating a new era in automation and artificial intelligence. Given the rapid advancements in robotics technology, its core area—robot control technology—has attracted increasing attention. Notably, sensors and sensor fusion technologies, which are considered essential for enhancing robot control technologies, have been widely and successfully applied in the field of robotics. Therefore, the integration of sensors and sensor fusion techniques with robot control technologies, which enables adaptation to various tasks in new situations, is emerging as a promising approach. This review seeks to delineate how sensors and sensor fusion technologies are combined with robot control technologies. It presents nine types of sensors used in robot control, discusses representative control methods, and summarizes their applications across various domains. Finally, this survey discusses existing challenges and potential future directions.

show abstract

“…(MLP) and Reflective Optical Sensors [89] Low cost, high accuracy and anti-interference ability, flexible architecture choice, suitable for embedded systems.…”

Section: Distance Estimation Methods Combining Multilayer Perceptronmentioning

confidence: 99%

Section: Infrared Sensorsmentioning

confidence: 99%

Robotics Perception and Control: Key Technologies and Applications

Luo,

Zhou,

Zeng

et al. 2024

Micromachines

View full text Add to dashboard Cite

show abstract

“…Measurement applications have had great success with the convolutional neural network (CNN), long short-term memory models [9,13,[16][17][18]. Analyzing the factors and proposing a suitable neural network model for measuring sensors are essential to improving sensor accuracy [19,20].…”

Section: Introductionmentioning

confidence: 99%

“…They require compensation factors to adapt different measuring voltage ranges [6][7][8]. Therefore, we have proposed an appropriate ANN model for the input factors that affect the measurement accuracy of the sensor [18,20,34]. Our test system includes a low-cost device distance sensor GP2Y0A02Y [35], and ESP32 [36] modules.…”

Section: Introductionmentioning

confidence: 99%

Improved accuracy of optical distance sensor based on artificial neural network applied to real-time systems

Bui¹,

Nguyen²,

Vu³

et al. 2022

Meas. Sci. Technol.

View full text Add to dashboard Cite

Optical time of flight (ToF) sensors has potentials in the revolution of distance measurement. These sensors can continously monitor the distance and track the object movement. However, the existed sensing methods for such distance optical sensors mainly calculate the flight time e.g., pulse transit and receive time without considering theenvironment effects. Therefore, the measurement accuracy is severely reduced. There areother technologies with higher accuracy in the distance measurement. Nonetheless, they are too expensive due to the high accurate power supply. In this paper, we innovatievely improve the accuracy in continous distance measurement using the artificial neural network (ANN) technique. The proposed method can be applied for very cheap optical distance sensors with analog output in a real-time system.. Moreover, the propose method can self-calibrate and be miniaturized for such cheap analog sensor applications. The prototype is built with the infrared sensor GP2Y0A02YK0F and Arduino control board (add the name of Adruino board), the ANN is implemented using the deep learning algorithm. The test results show that the distance measurement accurary is significantly improved and the measuring range is increased from 15 to 150 cm. Also, we calculate MSE, MAE, MBE, and R2 for further perfomance evaluation. The experimental results have proved the superior of the proposed ANN method in optical distance measurement. The proposed method can be applied for many types of sensors

show abstract

“…NNs are flexible model frameworks that "learn" patterns through an optimization algorithm that repeatedly iterates through the data. These models have countless realms of application, including predicting object distance, short-term rainfall forecasts, business financial distress, and chili plant disease classification (Mesa et al, 2019;Zhang et al, 2018;El Bannany et al, 2021;Nuanmeesri and Sriurai, 2021).…”

Section: Introductionmentioning

confidence: 99%

On the Use of Deep Neural Networks for Large-Scale Spatial Prediction

Gray¹,

Heaton²,

Bolintineanu³

et al. 2022

Journal of Data Science

View full text Add to dashboard Cite

For spatial kriging (prediction), the Gaussian process (GP) has been the go-to tool of spatial statisticians for decades. However, the GP is plagued by computational intractability, rendering it infeasible for use on large spatial data sets. Neural networks (NNs), on the other hand, have arisen as a flexible and computationally feasible approach for capturing nonlinear relationships. To date, however, NNs have only been scarcely used for problems in spatial statistics but their use is beginning to take root. In this work, we argue for equivalence between a NN and a GP and demonstrate how to implement NNs for kriging from large spatial data. We compare the computational efficacy and predictive power of NNs with that of GP approximations across a variety of big spatial Gaussian, non-Gaussian and binary data applications of up to size $n={10^{6}}$. Our results suggest that fully-connected NNs perform similarly to state-of-the-art, GP-approximated models for short-range predictions but can suffer for longer range predictions.

show abstract

Sensor Fusion for Distance Estimation Under Disturbance with Reflective Optical Sensors using Multi Layer Perceptron (MLP)

Cited by 5 publications

References 16 publications

Robotics Perception and Control: Key Technologies and Applications

Robotics Perception and Control: Key Technologies and Applications

Improved accuracy of optical distance sensor based on artificial neural network applied to real-time systems

On the Use of Deep Neural Networks for Large-Scale Spatial Prediction

Contact Info

Product

Resources

About