A high accuracy modeling scheme for dynamic systems: spacecraft reaction wheel model

Reaction wheels are crucial actuators in spacecraft attitude control subsystem (ACS). The precise modeling of reaction wheels is of fundamental need in spacecraft ACS for design, analysis, simulation, and fault diagnosis applications. The complex nature of the reaction wheel leads to modeling difficulties utilizing the conventional modeling schemes. Additionally, the absence of reaction wheel providers’ parameters is crucial for triggering a new modeling scheme. The Radial Basis Function Neural Network (RBFNN) has an efficient architecture, alluring generalization properties, invulnerability against noise, and amazing training capabilities. This research proposes a promising modeling scheme for the spacecraft reaction wheel utilizing RBFNN and an improved variant of the Quantum Behaved Particle Swarm Optimization (QPSO). The problem of enhancing the network parameters of the RBFNN at the training phase is formed as a nonlinear constrained optimization problem. Thus, it is proposed to efficiently resolve utilizing an enhanced version of QPSO with mutation strategy (EQPSO-2M). The proposed technique is compared with the conventional QPSO algorithm and different variants of PSO algorithms. Evaluation criteria rely upon convergence speed, mean best fitness value, stability, and the number of successful runs that has been utilized to assess the proposed approach. A non-parametric test is utilized to decide the critical contrast between the results of the proposed algorithm compared with different algorithms. The simulation results demonstrated that the training of the proposed RBFNN-based reaction wheel model with enhanced parameters by EQPSO-2M algorithm furnishes a superior prediction accuracy went with effective network architecture.

DEVELOPMENT OF INTELLIGENT SELF-BALANCING E-BIKE USING MACHINE LEARNING

The self-driving autonomous cars is becoming an increasingly popular concept all around the world but the area of self-driving two wheelers is still under developed. For developing countries like India, two wheelers are affordable than cars for most of the population. The project aims at developing intelligent self-balancing bike using artificial intelligence because the major problem in developing an autonomous bike is in the area of balancing. Even though there are many working mechanisms available for self-balancing of bike, the implementation of AI will be an edge over others from the point of computational power requirement and the programming complexity incurred. A prototype of the bike was developed with reaction wheel mechanism for self-balancing. The mechanism was fully controlled by AI by preventing the need of explicit programming for balancing which was the earlier technique used in self-balancing bike. Reinforcement learning, a type of machine learning technique is adopted for this purpose. The policy gradient algorithm was used to make the bike learn by itself for balancing. Even though the AI algorithm worked well in the virtual environment (balancing a cart-pole) it fails in the real environment. (i.e. it fails to balance the bike). It is because of the noisy data from the sensor, which gives inaccurate information about the orientation of the bike. The noise in the data is due to the vibration of the body when the reaction wheel rotates. This could be solved if the AI is fed with accurate information about the orientation of the vehicle.