Katherine Aro

Ms. Katherine Aro

orcid

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
3 December 2024 in Robotics
Verified
Journal Article
Open access
PDF file
KAKatherine Aro
Katherine Aro
Leonardo GuevaraLeonardo Guevara
Leonardo Guevara
MTMiguel Torres-Torriti
Miguel Torres-Torriti
Felipe A. TorresFelipe A. Torres
Felipe A. Torres
Alvaro Javier PradoAlvaro Javier Prado
Alvaro Javier Prado
This paper presents a robust control strategy for trajectory-tracking control of Skid-Steer Mobile Manipulators (SSMMs) using a Robust Nonlinear Model Predictive Control (R-NMPC) approach that minimises trajectory-tracking errors while overcoming model uncertainties and terra-mechanical disturbances. The proposed strategy is aimed at counteracting the effects of disturbances caused by the slip phenomena through the wheel–terrain contact and bidirectional interactions propagated by mechanical coupling between the SSMM base and arm. These interactions are modelled using a coupled nonlinear dynamic framework that integrates bounded uncertainties for the mobile base and arm joints. The model is developed based on principles of full-body energy balance and link torques. Then, a centralized control architecture integrates a nominal NMPC (disturbance-free) and ancillary controller based on Active Disturbance-Rejection Control (ADRC) to strengthen control robustness, operating the full system dynamics as a single robotic body. While the NMPC strategy is responsible for the trajectory-tracking control task, the ADRC leverages an Extended State Observer (ESO) to quantify the impact of external disturbances. Then, the ADRC is devoted to compensating for external disturbances and uncertainties stemming from the model mismatch between the nominal representation and the actual system response. Simulation and field experiments conducted on an assembled Pioneer 3P-AT base and Katana 6M180 robotic arm under terrain constraints demonstrate the effectiveness of the proposed method. Compared to non-robust controllers, the R-NMPC approach significantly reduced trajectory-tracking errors by 79.5% for mobile bases and 42.3% for robot arms. These results highlight the potential to enhance robust performance and resource efficiency in complex navigation conditions.

17 October 2024 in Preprints.org
Verified
Preprint
Open access
PDF file
KAKatherine Aro
Katherine Aro
Leonardo GuevaraLeonardo Guevara
Leonardo Guevara
Alvaro Javier PradoAlvaro Javier Prado
Alvaro Javier Prado
This paper presents a robust control strategy for trajectory tracking control of Skid-Steer Mobile Manipulators (SSMMs) using a Robust Nonlinear Model Predictive Control (R-NMPC) approach that minimizes trajectory tracking errors while overcoming model uncertainties and terra-mechanical disturbances. The proposed strategy aims to counteract disturbance effects caused by the slip phenomena through the wheel-terrain contact and bidirectional interactions propagated by the mechanical coupling between the SSMM base and arm. To capture such interactions, the mobile base dynamics and arm joints consider a coupled nonlinear model with bounded uncertainties based on principles of full-body energy balance and link torques. In addition, the SSMM dynamics stands by dynamical models, which are capable of capturing nonlinear speed relationships. Thus, a centralized control architecture integrates a nominal and ancillary controller based on the Active Disturbance Rejection Control (ADRC) to strengthen robustness of the Nonlinear Model Predictive Control (NMPC), operating the SSMM dynamics as a single robotic body. The NMPC is responsible for trajectory tracking control of a five Degrees of Freedom (DoF)’s SSMM. Meanwhile, the ADRC uses an Extended State Observer (ESO) to quantify the impact of external disturbances and uncertainties, generating compensation for unmeasurable disturbances. In addition, the proposed robust control strategy compensates for the model mismatch errors, which are generated by the difference between an SSMM nominal model (disturbance-free) and its real-time response. Results indicate that the approach is able to compensate for effects by 70% using pre-set trajectories and regulation tests against unmeasurable disturbances applied to simulated SSMR model.

22 July 2024 in Revista Iberoamericana de Automática e Informática Industrial
Verified
Journal Article
Open access
PDF file
SZSebastian Zapata
Sebastian Zapata
RURicardo Urvina
Ricardo Urvina
KAKatherine Aro
Katherine Aro
EAEduardo Aguilar
Eduardo Aguilar
FCFernando Auat Cheein
Fernando Auat Cheein
Alvaro Javier PradoAlvaro Javier Prado
Alvaro Javier Prado
Este trabajo introduce una estrategia híbrida de planificación de caminos para vehículos robóticos tipo diferencial, combinando métodos de aprendizaje por refuerzo con técnicas de muestreo aleatorio. Específicamente, se utiliza aprendizaje por refuerzo basado en Q-Learning (QL) para encontrar un camino global mediante la exploración y explotación de la información del entorno, donde un agente aprende a tomar acciones maximizando recompensas. El agente utiliza un método de muestreo RRT∗ (Rapidly-exploring Random Trees) para obtener puntos factibles de camino y acelerar la búsqueda, combinando las ventajas de QL con RRT∗ (MQL) para mejorar el muestreo y generar caminos suaves y factibles en espacios de alta dimensionalidad (Smooth Q-Learning - SMQL). Se realizó un análisis de rendimiento del método híbrido propuesto en condiciones de minería a cielo abierto, considerando criterios de maniobrabilidad, completitud, alcanzabilidad y robustez en entornos: i) rectos, ii) estrechos, iii) intrincados, y iv) tipo helicoidal con restricciones de terreno. Mediante simulaciones se demostró que SMQL supera las limitaciones de QL y RRT∗, al lograr mejor exploración y alcanzar r´ápida convergencia de recompensas. Por completitud, caminos previamente planificados con SMQL y MQL se prueban en un controlador de movimiento y un robot Husky A200, alcanzando una reducción del costo de error del 81.9 % y 76.4 % y esfuerzo de control del 79.8 % y 83.5 % en comparación con QL, respectivamente. Se espera que estos resultados tengan un impacto en el ahorro de recursos energéticos del robot al seguir rutas planificadas en ambientes mineros.

1 October 2018
Verified
Conference Paper
Closed access
KAKatherine Aro
Katherine Aro
NENemesis Escobar
Nemesis Escobar
Danilo ChavezDanilo Chavez
Danilo Chavez
PLPaulo Leica
Paulo Leica
Oscar Eduardo CamachoOscar Eduardo Camacho
Oscar Eduardo Camacho
This document developed a fuzzy supervisory control based on fuzzy logic for a four-tank multivariable system. A hierarchical control structure was established which basic principle is to decompose and coordinate highly complex systems and work them as subsystems. This hierarchical system was shaped by two layers. The first was a supervisory layer, in which initially the variable with major weight was found within the process to establish a hierarchy of variables, this was calculated through the decomposition into singular values of the Hankel matrix, also, considering the behavior of the plant in face of reference changes, it was sought to retune in line the parameters of the controllers arranged in the regulatory layer, this using fuzzy logic. The regulatory layer is the one that receives the parameters, which worked with independent control loops based on the structure of PID controllers that regulated the behavior of the four-tank system.