The study analyzes the robustness of implicit maps to external and internal disturbances in the task of simultaneous localization and mapping (SLAM) based on images from a depth camera (RGB-D). Using Co-SLAM as a representative method, we examined the sensitivity to three realistic perturbations: local bright spots and glare, inaccuracies in the calibration of camera intrinsics, and additive noise in the depth channel. For quantitative robustness evaluation, the root mean square error of the camera position is used after data alignment using the Kabsch-Umeyama algorithm. The results of the simulations with disturbances showed high robustness to local light flares, moderate to small inaccuracies in camera calibration, and low to noise in the depth channel. This study also proposes a new method for determining the initial position and orientation of the camera based on the linear and angular velocities from visual/visual-inertial odometry algorithm, which provides increased accuracy of the localization in the situation with noisy depth channel without a significant increase in computational complexity.

Motion planning for a group of mobile robots collaboratively transporting a load over non-uniform support surfaces is a challenging task. The core difficulty lies in simultaneously satisfying conflicting requirements: safe obstacle avoidance, maintaining group formation, minimizing travel time or path length, and adhering to dynamic constraints imposed by varying ground properties. This paper proposes a hybrid method that overcomes the fundamental limitations of existing approaches, such as the low computational speed of sampling-based methods (RRT*), the non-strict constraint satisfaction in reinforcement learning (RL), and the rigid smoothness requirements of classical optimal control methods. The key idea is to combine optimal control with machine learning, where a neural network model approximates the objective function — a more versatile approach compared to approximating system dynamics or environment models. Two training strategies are investigated: supervised learning for accurately reproducing an objective function and reinforcement learning (DDPG algorithm) for flexibly defining it via a reward function. Constraints, including acceleration limits dependent on the surface type, are defined analytically using a smoothed discrete space and bilinear interpolation. The optimization problem is solved using the high-performance solver FATROP, integrated with the CasADi automatic differentiation framework. Simulation results demonstrate that the proposed method outperforms a previous RRT*-based implementation, reducing trajectory computation time 58-fold and decreasing path traversal time by 20 %. The experiments also showed that reinforcement learning allows for flexibly redefining the optimization goal, finding a shorter path (8.9 m vs. 9.4 m) at the cost of a slight increase in travel time. The FATROP solver also proved to be 22 % faster than the traditional IPOPT. These results confirm the potential of the hybrid approach for multi-robot motion planning tasks, enabling the integration of complex nonlinear dependencies without a significant loss in performance.

The classical linear programming transportation problem of minimizing the cost of transportation between production and consumption points has many applications, one of which is the problem of efficient fire control. The article considers a modified formulation of this problem. It includes main and auxiliary batteries, each of which can fire a limited number of shots at targets with a given efficiency. Auxiliary batteries can fire only at predetermined targets. The goal is to distribute targets between batteries in such a way that the total efficiency of destruction is maximized. The decomposition method is used to solve the proposed problem. The original problem of large dimension is divided into many simpler one-dimensional and two-dimensional subproblems. At the first stage, the initial pseudo-solution is found as a set of solutions to these subproblems. If it is admissible for the original problem, then it is also optimal. Otherwise, an iterative process of sequentially coordinating the solutions of the subproblems is launched by cyclically recalculating the coefficients of the objective function in two dimensional problems. This process guarantees a monotonic approximation to the optimal solution. The article examines in detail possible cases arising during the algorithm operation, including a special degenerate case, for the resolution of which it is proposed to introduce additional constraints. The possibility of replacing inequality constraints with equality constraints for the main batteries within the framework of the decomposition approach without loss of generality is theoretically substantiated. The efficiency of the proposed algorithm is confirmed by the results of computational experiments. Approximation of the dependence of the running time on the problem dimension demonstrates the polynomial complexity of the method. The obtained results open up prospects for applying this approach to other non classical formulations of transport-type optimization problems.

In recent years, the popularity of small multirotor unmanned aerial vehicles and, in particular, quadrocopters (QC) has increased significantly, due to both their characteristics and the low cost of manufacturing and operation. At the same time, one of the promising ways to increase the efficiency of using QC to solve a wide variety of tasks in both civilian and military spheres is to improve the flight control algorithms of the spacecraft. However, despite the presence of numerous classical and modern methods for synthesizing flight control algorithms, many of them turn out to be ineffective in conditions of a priori uncertainty of the mathematical model of vehicle dynamics, as well as in conditions of wind disturbance. QC as an object of control is a complex, nonlinear, multidimensional, multi-connected dynamic system of the 12th order with the presence of indeterminate parameters and external disturbances. The paper considers the problem of synthesizing an algorithm for the subsequent flight control of a spacecraft, which provides tracking by a vector of controlled variables of its arbitrarily set, programmatic change. As indicators of the effectiveness of the synthesized monitoring control algorithm, the QC summer uses, firstly, direct indicators of the quality of the control process (control time and overshoot), characterizing the speed and tendency of the system to oscillate, and, secondly, the maximum amplitude of the controlling effects, characterizing the energy consumption for their generation. The article is devoted to a comparative analysis of the parametric robustness properties of QC flight tracking algorithms synthesized on the basis of the most popular modern methods of nonlinear control of dynamic objects: the sliding mode method, the integrator bypass method, the feedback linearization method, the MPC proactive control method, the "deep" feedback method, the method of inverse dynamics problems with compensation for non-linearity. At the same time, the effectiveness of the synthesized flight tracking control algorithms was analyzed by computer verification in the Python environment under conditions of variable load and uncontrolled disturbing wind effects.

Currently, the integration of satellite navigation systems (SNS) and correlation-extremal navigation systems (CENS) for unmanned vehicles (UVs) is implemented based on principles of separate or, at best, weakly coupled integration, where their measurements are processed by different navigation algorithms (stochastic filters) with subsequent correction of inevitable discrepancies using various optimization methods. This approach is characterized by both high computational costs due to the need for parallel implementation of SNS and CENS measurement processing algorithms and subsequent optimization problem solving, as well as critical dependence of positioning accuracy on increasing levels of radio measurement interference. In this regard, a solution is proposed to improve the positioning accuracy of UVs based on the principle of tightly coupled integration, which involves representing the UV’s coordinate vector and the terrain elevation of the underlying surface as a single navigation vector, estimated by a common stochastic filter. Such measurement processing, in addition to significantly reducing computational costs, ensures robust and high-precision estimation of UV navigation parameters under conditions of intense interference of both natural and artificial origin. The results of a numerical experiment illustrating the effectiveness of the proposed approach are presented.

The object of research in the article is the information technology of autonomous group control of a multi-satellite system for remote sensing of the Earth. An orbital grouping of small satellite clusters is considered as case study. Clusters refer to satellites that are located close to each other. The information technology is based on an agent-oriented approach and information interaction of satellite software agents based on the use of inter-satellite communications. The information interaction of agents is considered taking into account the dynamics of the establishment of inter-satellite communication lines over time. It is assumed that communication within clusters is possible in real time, and communication between clusters is only possible within time intervals during which the necessary conditions are met. The purpose of information interaction protocols is the autonomous group solution (without the participation of the ground control complex) of the following tasks: 1) distribution and redistribution of requests within each cluster and 2) between clusters, 3) determination and coordination of the procedure for transmitting observation data to Earth during the establishment of communication sessions with ground points, and 4) search for an appropriate distribution of survey data between clusters using inter-satellite communication, ensuring a reduction in the time of delivery of survey data to Earth. In the process of participating in information interaction protocols, agents perform autonomous planning of the targeted use of their satellite. This planning includes finding an acceptable shooting plan and forming a flight plan taking into account the technical capabilities and limitations of the satellite, as well as taking into account the control of the electrical balance. Demonstration of capabilities and evaluation of the effectiveness of information technology is carried out using software implementation of its simulation model. The article provides a description of examples of output data that are generated during simulation modeling.