Roman Lavrenov

Comparing fiducial marker systems in the presence of occlusion

A fiducial marker system is a system of unique 2D (planar) marker, which is placed in an environment and automatically will be detected with a camera with a help of a corresponding detection algorithm. Application areas of these markers include industrial systems, augmented reality, robots navigation, human-robot interaction and others. Marker system designed for such different applications must be robust to such factors as view angles, occlusions, changing distances, etc. This paper compares three existing systems of markers: ARTag, AprilTag, and CALTag. As a benchmark, we use their reliability and detection rate in presence of occlusions of various types and intensity. The paper presents experimental comparison of these markers. The marker detection was performed with a simple inexpensive Web camera.

Modified Spline-based Path Planning for Autonomous Ground Vehicle.

Potential function based methods play significant role in global and local path planning. While these methods are characterized with good reactive behavior and implementation simplicity, they suffer from a well-known problem of getting stuck in local minima of a navigation function. In this article we propose a modification of our original spline-based path planning algorithm for a mobile robot navigation, which succeeds to solve local minima problem and adds additional criteria of start and target points visibility to help optimizing the path selection. We apply a Voronoi graph based path as an input for iterative multi criteria optimization algorithm. The algorithm was implemented in Matlab environment and simulation results demonstrate that we succeeded to overcome our original algorithm pitfalls.

Modified Spline-Based Navigation: Guaranteed Safety for Obstacle Avoidance

Successful interactive collaboration with a human demands mobile robots to have an advanced level of autonomy, which basic requirements include social interaction, real time path planning and navigation in dynamic environment. For mobile robot path planning, potential function based methods provide classical yet powerful solutions. They are characterized with reactive local obstacle avoidance and implementation simplicity, but suffer from navigation function local minima. In this paper we propose a modification of our original spline-based path planning algorithm, which consists of two levels of planning. At the first level, Voronoi-based approach provides a number sub-optimal paths in different homotopic groups. At the second, these paths are optimized in an iterative manner with regard to selected criteria weights. A new safety criterion is integrated into both levels of path planning to guarantee path safety, while further optimization of a safe path relatively to other criteria is secondary. The modified algorithm was implemented in Matlab environment and demonstrated significant advantages over the original algorithm.

Voronoi-based trajectory optimization for UGV path planning

Optimal path planning in dynamic environments for an unmanned vehicle is a complex task of mobile robotics that requires an integrated approach. This paper describes a path planning algorithm, which allows to build a preliminary motion trajectory using global information about environment, and then dynamically adjust the path in real-time by varying objective function weights. We introduce a set of key parameters for path optimization and the algorithm implementation in MATLAB. The developed algorithm is suitable for fast and robust trajectory tuning to a dynamically changing environment and is capable to provide efficient planning for mobile robots.