Abstract. The application of mobile robots is becoming more and more common in order to perform various tasks under conditions where the presence of the human in their site is impossible on safety grounds or unacceptable due to the reduction of the productivity of the technological equipment served by them. The objective of robot movement planning is to guarantee the desirable robot movement path as it moves across the planned path on the basis of controlling impacts, i.e. sensors. Numerous investigations were carried out under unknown environment conditions, which were intended for addressing the problem of robot movement without colliding with obstacles in its path by employing various navigation methods. The purpose of this paper is to analyse navigation methods employed for addressing the problem of robot movement without colliding with obstacles in its path under unknown environment conditions. The paper analyses the multi-agent system framework, generalized agent control system framework, local path planning algorithm for robots in an unknown two-dimensional space and other related questions in order to reach the above-mentioned purpose.Keywords: geometric path planning method, multi-agent system, artificial potential method, global planning, local planning.
IntroductionThe development of mobile robotics is determined by the desire to replace men at hard and routine, dangerous and responsible work, i.e. in warehouses for efficiently moving materials from stocking shelves to order fulfilment zones, in hospitals for moving materials, in military for clearing mines, in household for vacuuming or gardening etc. The most important characteristics of such robot are mobility and autonomy. In this case, autonomy is necessary in order to guarantee the movement in an unknown environment so that the mobile robot (MR) is able to independently move in an unknown environment performing the task set.The mobile robot movement planning is the most important issue of the autonomous robototechnical system control and one of the fields of modern science and practice most actively investigated. The solution to the task of robot movement planning colliding encompasses questions related with scientific fields such as artificial intelligence, computational geometry, computer simulation and the theory of automatic control. The automation of the movement planning process by also minimising the time costs for preparation and completion operations as well as by accelerating the process of switching the robot from one production task to another constitute the basis for flexible production organisation.Various achievements of scientists in this field enabled to determine the priority trends of mobile robot development. The main design tasks of mobile robot movement in unknown environment conditions are provided in Figure 1 [1; 2].It is easy to understand that the solution to the tasks listed above (Figure 1) requires the solution to a variety of main autonomous mobile robot tasks. First, the robot shall plan a fast and passable route i...