TY  - JOUR
AU  - Singh, Nidhi
AU  - Nigam, Prashant
AU  - Dewan, Lillie
PY  - 2026
DA  - 2026/11/02
TI  - Genetic Algorithm Based PFOID-SMC with a Sliding Mode Observer for a Quadcopter under Sensing and Communication Constraints
JO  - ICCK Transactions on Sensing, Communication, and Control
T2  - ICCK Transactions on Sensing, Communication, and Control
JF  - ICCK Transactions on Sensing, Communication, and Control
VL  - 3
IS  - 4
SP  - 210
EP  - 236
DO  - 10.62762/TSCC.2026.637136
UR  - https://www.icck.org/article/abs/TSCC.2026.637136
KW  - Genetic algorithm
KW  - fractional-order control
KW  - sliding mode control
KW  - sliding mode observer
KW  - parametric uncertainty
KW  - unmanned aerial vehicle
KW  - sensor modelling
KW  - networked control
AB  - This paper presents a genetic algorithm (GA)-tuned sliding mode controller for a quadcopter (QuCo) in which the sliding surface is built from a proportional-fractional-order-integral-derivative (PFOID) error function (PFOID-SMC). The aim is to achieve accurate trajectory tracking in the presence of aerodynamic drag, wind, sensor noise and parametric uncertainty. Sensing and communication are kept inside the closed loop. The controller receives only measured or estimated states from a gyroscope (200~Hz), an attitude and heading reference system (100~Hz), an RTK-GNSS receiver (10~Hz) and a barometer (50~Hz), each with its own bias and noise, while the reference and the GNSS position are sent over a link with delay, packet loss and quantization. A sliding mode observer reconstructs attitude, angular rates, velocities, wind and disturbance accelerations and motor faults, and its disturbance estimates enter the equivalent control. Finite-time reaching and exponential convergence of the sliding variables are proved channel by channel with Lyapunov functions, and the stability of the complete cascade is checked numerically. The controller is compared with GA-tuned PID, conventional SMC, FOPID, super-twisting SMC and the earlier PFOID-SMC law in 15 scenarios, each with five noise realizations. In nominal conditions, it is not the most accurate (3-D position RMSE 0.119~m, against 0.071~m for FOPID). It gives the lowest error, however, under the strongest wind (0.134~m), under a 30% increase in mass and inertia (0.179~m, against at least 0.82~m for all baselines) and after a 40% loss of effectiveness of one motor (0.147~m). Without the observer, the same controller diverges.
SN  - 3068-9287
PB  - Institute of Central Computation and Knowledge
LA  - English
ER  - 

@article{Singh2026Genetic,
  author = {Nidhi Singh and Prashant Nigam and Lillie Dewan},
  title = {Genetic Algorithm Based PFOID-SMC with a Sliding Mode Observer for a Quadcopter under Sensing and Communication Constraints},
  journal = {ICCK Transactions on Sensing, Communication, and Control},
  year = {2026},
  volume = {3},
  number = {4},
  pages = {210-236},
  doi = {10.62762/TSCC.2026.637136},
  url = {https://www.icck.org/article/abs/TSCC.2026.637136},
  abstract = {This paper presents a genetic algorithm (GA)-tuned sliding mode controller for a quadcopter (QuCo) in which the sliding surface is built from a proportional-fractional-order-integral-derivative (PFOID) error function (PFOID-SMC). The aim is to achieve accurate trajectory tracking in the presence of aerodynamic drag, wind, sensor noise and parametric uncertainty. Sensing and communication are kept inside the closed loop. The controller receives only measured or estimated states from a gyroscope (200~Hz), an attitude and heading reference system (100~Hz), an RTK-GNSS receiver (10~Hz) and a barometer (50~Hz), each with its own bias and noise, while the reference and the GNSS position are sent over a link with delay, packet loss and quantization. A sliding mode observer reconstructs attitude, angular rates, velocities, wind and disturbance accelerations and motor faults, and its disturbance estimates enter the equivalent control. Finite-time reaching and exponential convergence of the sliding variables are proved channel by channel with Lyapunov functions, and the stability of the complete cascade is checked numerically. The controller is compared with GA-tuned PID, conventional SMC, FOPID, super-twisting SMC and the earlier PFOID-SMC law in 15 scenarios, each with five noise realizations. In nominal conditions, it is not the most accurate (3-D position RMSE 0.119~m, against 0.071~m for FOPID). It gives the lowest error, however, under the strongest wind (0.134~m), under a 30\% increase in mass and inertia (0.179~m, against at least 0.82~m for all baselines) and after a 40\% loss of effectiveness of one motor (0.147~m). Without the observer, the same controller diverges.},
  keywords = {Genetic algorithm, fractional-order control, sliding mode control, sliding mode observer, parametric uncertainty, unmanned aerial vehicle, sensor modelling, networked control},
  issn = {3068-9287},
  publisher = {Institute of Central Computation and Knowledge}
}