Neuro-predictive control system for a mobile nonholonomic three-wheeled robot in an environment with static obstacles

This article proposes to track and predict the trajectory of an autonomous nonholonomic three-wheeled mobile robot in an environment with static obstacles using a neuro-predictive control system. This system consists of a modified Elman neural network (to track the position and orientation of the robot), a neural network model of an obstacle (to determine the point cloud of an obstacle) and cubic spline curve interpolation methods and a PSO algorithm (to smooth the obstacle avoidance curve and ensure the shortest distance). A new trajectory for avoiding an obstacle is built on three points (before the obstacle, the center of the obstacle, after the obstacle). The proposed control system improves the efficiency of mobile robot control and provides the smallest deviation from the movement trajectory, in general, and in the place where the obstacle is bypassed, in particular. The neuro-predictive control system is compared with the classical PSO algorithm, and, within the system itself, methods for smoothing the obstacle avoidance curve (cubic spline interpolation and PSO algorithm) are compared. Algorithms are compared according to such criteria as the average distance of the robot from the obstacle when rebuilding the trajectory, the speed of movement, the time it takes to bypass the obstacle. In addition, the deviation from the given trajectory of movement is checked: movements along the lemniscate and along the square. The simulation results showed that the neuro-predictive system is more efficient (by 28.1 % on average) in avoiding an obstacle (provides the shortest distance) and performs this maneuver faster (by 17.2 % on average) than the classical PSO algorithm. Also, within the system itself, the PSO-algorithm works more efficiently to construct an obstacle avoidance curve (3.3 % closer to the obstacle and, on average, 88.2 % less root-mean-square error) than cubic spline interpolation. At the same time, the neuro-predictive control system copes much better with following the desired trajectory than the classical PSO algorithm.

1. Su K.H., Chen Y.Y., Su S.F. Design of neural-fuzzy-based controller for two autonomously driven wheeled robot. Neurocompting. 2015;73:2478–2488.

2. Berezina V.A., Mezentseva O.S., Mezentsev D.V. Modified Elman neural network with dynamic learning rate for tracking and motion prediction of a nonholonomic three-wheeled mobile robot. Modeling, Optimization and Information Technology. 2022;10(3). Available from: https://moitvivt.ru/ru/journal/pdf?id=1199 DOI: 10.26102/2310-6018/2022.38.3.003 (In Russ.) (accessed on 01.11.2022).

3. Lucas R., Oliveira R.M., Nascimento C.B., Kaster M.S. Performance analysis of an adaptive Gaussian nonlinear PID control applied to a step-down CC-CC converter. IEEE 24th International Symposium on Industrial Electronics (ISIE). 2015;1:1022–1026.

4. Li Z., Yang C., Su C., Deng J., Zhang W. Vision-Based Model Predictive Control for Steering of a Nonholonomic Mobile Robot. IEEE Transactions on Control Systems Technology. 2019;24(2):553–564.

5. Suprapto B. Y., Mustaqim A., Wahab W., Kusumoputro B. Modified elman recurrent neural network for attitude and altitude control of heavy-lift hexacopter. 15th International Conference on Quality in Research (QiR): International Symposium on Electrical and Computer Engineering. 2017;1:309–314.

6. He W. Adaptive neural network control of an uncertain robot with fullstate constraints. IEEE Transactions on Cybernetics. 2016;46(3):620–629.

7. Al-Jamali N.A., Al-Raweshidy H.S. Modified Elman Spike Neural Network for Identification and Control of Dynamic System. IEEE Access. 2020;8:61246–61254.

8. Li D., Liu Y., Tong S., Chen C.L. Approximation-Based Adaptive Neural Tracking Control of Nonlinear MIMO Unknown Time-Varying Delay Systems with Full State Constraints. IEEE Transactions on Cybernetics. 2017;47(10):3100–3109.

9. Deng L., Seethaler R. J., Chen Y., Yang P., Cheng Q. Modified Elman neural network based neural adaptive inverse control of rate-dependent hysteresis. International Joint Conference on Neural Networks (IJCNN). 2016;1:2366–2373.

10. Wang S., Zhang Y., Dong Z., Du S., Ji G. Feed-forward neural network optimized by hybridization of PSO and ABC for abnormal brain detection. International Journal of Imaging Systems & Technology. 2015;25(2):153–164.