Robot path planning problem is one of most important task mobile robots. This paper proposes an original approach using a description by string cubic splines. Such easy executable and natural for car-like robot. Furthermore, it possible to ensure smooth derivation in connections particular In this case, the equivalent optimization parameters An evolutionary technique called particle swarm (PSO) was used hereunder due its relatively fast convergence global search character. Various settings...

An approach for coordination and control of 3D heterogeneous formations unmanned aerial ground vehicles under hawk-eye-like relative localization is presented in this paper. The core the method lies use visual top-view feedback from flying robots stabilization entire group a leader–follower formation. We formulate novel model predictive control-based methodology guiding employed to solve trajectory planning virtual leader into desired target region. In addition, used keeping following shape...

In this paper, we present a small, light-weight, low-cost, fast and reliable system designed to satisfy requirements of relative localization within swarm micro aerial vehicles. The core the proposed solution is based on off-the-shelf components consisting Caspa camera module Gumstix Overo board accompanied by developed efficient image processing method for detecting black white circular patterns. Although idea roundel recognition simple, exhibits estimation position pattern up 30 fps using...

A stabilization and control technique developed for steering swarms of unmanned micro aerial vehicles is proposed in this paper. The presented approach based on a visual relative localization swarm particles designed utilization multi-robot teams real-world dynamic environments. core the swarming behaviour inspired by Reynold's BOID model 2D simulations schooling fish. idea simple model, with three rules: Separation, Alignment Cohesion, extended quadrotors solution integrates mechanism,...

We present an implementation of the Speeded Up Robust Features (SURF) on a Field Programmable Gate Array (FPGA). The SURF algorithm extracts salient points from image and computes descriptors their surroundings that are invariant to scale, rotation illumination changes. interest point detection feature descriptor extraction is often used as first stage in autonomous robot navigation, object recognition tracking etc. However, computationally demanding therefore can't be systems with limited...

E-learning is a modern and effective approach for training in various areas at different levels of education. This paper gives an overview SyRoTek, e-learning platform mobile robotics, artificial intelligence, control engineering, related domains. SyRoTek provides remote access to set fully autonomous robots placed restricted area with dynamically reconfigurable obstacles, which enables solving huge variety problems. A user able the real time by their own developed algorithms as well being...

Abstract This article describes a simple monocular navigation system for mobile robot based on the map‐and‐replay technique. The presented method is robust and easy to implement does not require sensor calibration or structured environment, its computational complexity independent of environment size. can navigate while sensing only one landmark at time, making it more than other approaches. aforementioned properties allow even low‐cost robots effectively act in large outdoor indoor...

In this paper, we discuss the problem of goal assignment in multi-robot exploration task. The presented work is focused on underlying optimal task allocation that addressed by three state-of-the art approaches. addition, propose a novel strategy considering all current goals (not only immediate goal) for each robot, which leads to multiple traveling salesman formulation. Although strongly NP-hard, show its approximate solution computationally feasible and overall requirements are competitive...

Inspection systems utilizing unmanned aerial vehicles (UAVs) equipped with thermal cameras are increasingly popular for the maintenance of photovoltaic (PV) power plants. However, automation inspection task is a challenging problem as it requires precise navigation to capture images from optimal distances and viewing angles. This paper presents novel localization pipeline that directly integrates PV module detection UAV navigation, allowing positioning during inspection. Detections used...

Accurate models of the environment are a crucial requirement for autonomous mobile robots. The process how to acquire knowledge about operating is one most challenging problems in this research area. quality model depends on number and types sensors used. Occupancy grids common low-level used robotics fusion noisy data. This paper first introduces novel method building an occupancy grid from monocular color camera. next part work describes camera data with rangefinder. final presents new...

We present a simple and robust monocular camera-based navigation system for an autonomous quadcopter. The method does not require any additional infrastructure like radio beacons, artificial landmarks or GPS can be easily combined with other methods algorithms. Its computational complexity is independent of the environment size it works even when sensing only one landmark at time, allowing its operation in poor environments. also describe FPGA based embedded realization method's most...

Performance of exploration strategies strongly depends on the process determination a next robot goal. Current approaches define different utility functions how to evaluate and select possible goal candidates. One mostly used evaluation criteria is distance cost that prefers candidates close current position. If this only criterion, simply nearest candidate chosen as Although criterion simple implement gives feasible results there are situations where leads wrong decisions. This paper...

A navigation and stabilization scheme for 3D heterogeneous (UAVs UGVs) formations acting under a hawk-eye like relative localization is presented in this paper. We formulate novel Model Predictive Control (MPC) based concept formation driving leader-follower constellation into required target region. The to region problem solved using the MPC methodology both: i) trajectory planning control of virtual leader, ii) followers - UAVs UGVs. core method lies avoidance function on model respecting...

We revise, improve and extend the system previously introduced by us named SSM-VPR (Semantic Spatial Matching Visual Place Recognition), largely boosting its performance above current state of art. The encodes images places employing activations different layers a pre-trained, off-the-shelf, VGG16 Convolutional Neural Network (CNN) architecture. It consists two stages: given query image place, (1) list candidates is selected from database (2) are geometrically compared with query. comparison...

We present a fast and precise vision-based software intended for multiple robot localization. The core component of the proposed localization system is an efficient method black white circular pattern detection. robust to variable lighting conditions, achieves sub-pixel precision, its computational complexity independent processed image size. With off-the-shelf equipment low-cost camera, algorithm able process hundreds images per second while tracking objects with millimeter precision....

The ability to move in complex environments is a key property required for deployment of modular robots challenging applications like search & rescue missions or space exploration. Wide range motion types crawling walking can be achieved using Central Pattern Generators producing periodic control signals. Although these motions very effective steer their vicinity given direction, they need switched reach far position the environment. This paper presents novel modification Rapidly Exploring...

We present a Visual Place Recognition system that follows the two-stage format common to image retrieval pipelines. The encodes images of places by employing activations different layers pre-trained, off-the-shelf, VGG16 Convolutional Neural Network (CNN) architecture. In first stage our method and given query place, number top candidate is retrieved from previously stored database places. second stage, we propose an exhaustive comparison against these candidates encoding semantic spatial...