Studying the impact of the optimal control problem components on its frequency for a quadruple walking robot

There exist a large number of classes of the walking robots differing in the number of limbs, principles and modes of motion and control. Each class of robots has its own advantages and disadvantages depending on the tasks to be solved. A quadruple walking robot is considered capable of moving over the rough terrain. Its advantages over the biped systems include higher stability, and over the six- and eight-legged systems - simpler control due to the smaller number of limbs, and the possibility of using animals as the prototypes. One of the important and not completely solved problems is development of approaches to control a quadruple walking robot, as well as establishing the influence of computational complexity of the control system separate elements on its overall operation frequency. Such robot control system is represented in the form of three modules: inverse kinematics, predictive control and model parameters calculation. Each module requires solving a certain problem. A study was conducted on the time distribution between the control system modules in a single iteration. Average time distribution values and their standard deviations were determined for the two robot motion modes: standing and walking. Results of the numerical experiments established that the standard deviations were not depending on the device motion mode, which indicated independence of the computational load created by the control system on the robot motion mode.

References

[1] Katayama S., Ohtsuka T. Whole-body model predictive control with rigid contacts via online switching time optimization. IEEE IROS, 2022, doi: https://doi.org/10.1109/IROS47612.2022.9981790

[2] Bledt G., Wensing P.M., Ingersoll S. et al. Contact model fusion for event-based locomotion in unstructured terrains. IEEE ICRA, 2018, pp. 4399–4406, doi: https://doi.org/10.1109/ICRA.2018.8460904

[3] Kim D., Di Carlo J., Katz B. et al. Highly dynamic quadruped locomotion via whole-body impulse control and model predictive control. arXiv, doi: https://doi.org/10.48550/arXiv.1909.06586

[4] Ding Y., Pandala A., Li C., Shin Y.H., Park H.W. Representation-free model predictive control for dynamic motions in quadrupeds. IEEE Trans. Robot., 2021, vol. 37, no. 4, pp. 1154–1171, doi: https://doi.org/10.1109/TRO.2020.3046415

[5] Mason S., Rotella N., Schaal S. et al. Balancing and walking using full dynamics LQR control with contact constraints. IEEE-RAS Humanoids, 2016, pp. 63–68, doi: https://doi.org/10.1109/HUMANOIDS.2016.7803255

[6] Savin S., Balakhnov O., Khusainov R. et al. State observer for linear systems with explicit constraints: orthogonal decomposition method. Sensors, 2021, vol. 21, no. 18, art. 6312, doi: https://doi.org/10.3390/s21186312

[7] Manko S.V., Shestakov E.I. Automatic synthesis of gait scenarios for reconfigurable mechatronic modular robots in the modification of the walking platform. Russian Technological Journal, 2018, vol. 6, no. 4, pp. 26–41, doi: https://doi.org/10.32362/2500-316X-2018-6-4-26-41 (in Russ.).

[8] Vasilyanov G.S., Lavrov A.A. Development and research of algorithms of adaptive management of the quadruped robotized platforms. Ekstremalnaya robototekhnika [Extreme Robotics], 2015, vol. 1, no. 1, pp. 323–332. (In Russ.).

[9] Varlashin V.V., Ivanov A.A., Rogov A.V. et al. Modeling the gaits of a walking robot on a plane. Ekstremalnaya robototekhnika [Extreme Robotics], 2016, vol. 1, no. 1, pp. 149–153. (In Russ.).

[10] Teteryatnikov A.G., Rybakov A.V., Vybornov N.A. et al. Simulation of a four-legged walking robot stabilization system. Robototekhnika i tekhnicheskaya kibernetika [Robotics and Technical Cybernetics], 2020, vol. 8, no. 3, pp. 189–197, doi: https://doi.org/10.31776/RTCJ.8304 (in Russ.).

[11] Dobrynin D.A. [Modelling of learning control system of a walking robot]. Shestoy Vseros. nauch.-prakt. seminar Bespilotnye transportnye sredstva s elementami iskusstvennogo intellekta [6th Russ. Int. Sci.-Pract. Seminar Unmanned Vehicles with Artificial Intelligence Elements]. Moscow, RAII Publ., 2021, pp. 109–111. (In Russ.).

[12] Danilov V.A., Diane S.A.K. [Modelling of adaptive behaviour of multilink walking robots]. Shestoy Vseros. nauch.-prakt. seminar Bespilotnye transportnye sredstva s elementami iskusstvennogo intellekta [6th Russ. Int. Sci.-Pract. Seminar Unmanned Vehicles with Artificial Intelligence Elements]. Moscow, RAII Publ., 2021, pp. 132–134. (In Russ.).

[13] Tszun S. Research on control strategies for four-legged walking robots. StudNet, 2021, vol. 4, no. 7, pp. 557–572. URL: https://stud.net.ru/wp-content/uploads/2021/07/%D0%92%D1%8B%D0%BF%D1%83%D1%81%D0%BA-7_2021-1.pdf (in Russ.).

[14] Karginov L.A. The synthesis of control actions to drive the actuators of walking robots without solving the inverse kinematics task. Inzhenernyy zhurnal: nauka i innovatsii [Engineering Journal: Science and Innovation], 2013, no. 4, doi: http://dx.doi.org/10.18698/2308-6033-2013-4-692 (in Russ.).

[15] Borisov I.I., Kolyubin S.A. Energy-based analysis of bioinspired mechanism for cheetah robot leg. Nauchno-tekhnicheskiy vestnik informatsionnykh tekhnologiy, mekhaniki i optiki [Scientific and Technical Journal of Information Technologies, Mechanics and Optics], 2019, vol. 19, no. 6, pp. 1106–1114, doi: https://doi.org/10.17586/2226-1494-2019-19-6-1106-1114 (in Russ.).

[16] Di Carlo J., Wensing P.M., Katz B. et al. Dynamic locomotion in the mit cheetah 3 through convex model-predictive control. IEEE/RSJ IROS, 2018, doi: https://doi.org/10.1109/IROS.2018.8594448

[17] Baerlocher P., Boulic R. Task-priority formulations for the kinematic control of highly redundant articulated structures. IEEE/RSJ Int. Conf. on Intelligent Robots and Systems. Innovations in Theory, Practice and Applications, 1998, vol. 1, pp. 323–329, doi: https://doi.org/10.1109/IROS.1998.724639