Mariusz Wzorek

Reconfigurable Path Planning for an Autonomous Unmanned Aerial Vehicle

In this paper, we present a motion planning framework for a fully deployed autonomous unmanned aerial vehicle which integrates two sample-based motion planning techniques, Probabilistic Roadmaps and Rapidly Exploring Random Trees. Additionally, we incorporate dynamic reconfigurability into the framework by integrating the motion planners with the control kernel of the UAV in a novel manner with little modification to the original algorithms. The framework has been verified through simulation and in actual flight. Empirical results show that these techniques used with such a framework offer a surprisingly efficient method for dynamically reconfiguring a motion plan based on unforeseen contingencies which may arise during the execution of a plan. The framework is generic and can be used for additional platforms.

Micro Unmanned Aerial Vehicle Visual Servoing for Cooperative Indoor Exploration

Recent advances in the field of micro unmanned aerial vehicles (MAVs) make flying robots of small dimensions suitable platforms for performing advanced indoor missions. In order to achieve autonomous indoor flight a pose estimation technique is necessary. This paper presents a complete system which incorporates a vision-based pose estimation method to allow a MAV to navigate in indoor environments in cooperation with a ground robot. The pose estimation technique uses a lightweight light emitting diode (LED) cube structure as a pattern attached to a MAV. The pattern is observed by a ground robots camera which provides the flying robot with the estimate of its pose. The system is not confined to a single location and allows for cooperative exploration of unknown environments. It is suitable for performing missions of a search and rescue nature where a MAV extends the range of sensors of the ground robot. The performance of the pose estimation technique and the complete system is presented and experimental flights of a vertical take-off and landing (VTOL) MAV are described.

Control System Framework for Autonomous Robots Based on Extended State Machines

We present a new framework optimized for the design, implementation, and testing of control systems for autonomous robots. It is based on a new visual specification language which specifies both control and data flow, and which is suited to be interpreted in real-time. The framework is divided into a comprehensive development and a lightweight run-time environment. The latter is fully integrated with a real-time operating system and permits to reconfigure a control system without compilation at run time. Moreover, communication in distributed systems is supported. An earlier version of the framework has been successfully applied in an autonomous helicopter and an autonomous ground vehicle project

Vision-based pose estimation for autonomous indoor navigation of micro-scale Unmanned Aircraft Systems

We present a navigation system for autonomous indoor flight of micro-scale Unmanned Aircraft Systems (UAS) which is based on a method for accurate monocular vision pose estimation. The method makes use of low cost artificial landmarks placed in the environment and allows for fully autonomous flight with all computation done on-board a UAS on COTS hardware. We provide a detailed description of all system components along with an accuracy evaluation and a time profiling result for the pose estimation method. Additionally, we show how the system is integrated with an existing micro-scale UAS and provide results of experimental autonomous flight tests. To our knowledge, this system is one of the first to allow for complete closed-loop control and goal-driven navigation of a micro-scale UAS in an indoor setting without requiring connection to any external entities.

High Accuracy Ground Target Geo-location Using Autonomous Micro Aerial Vehicle Platforms

This paper presents a method for high accuracy ground target localization using a Micro Aerial Vehicle (MAV) equipped with a video camera sensor. The proposed method is based on a satellite or aeri ...

LINKMAV, A PROTOYPE ROTARY WING MICRO AERIAL VEHICLE

Abstract This paper provides an overview about the design of the LinkMAV rotary wing autonomous Micro Aerial Vehicle. The paper describes the flying platform, the main problems related to aerodynamics and propulsion, the onboard avionics and the flight control system, including the autonomous navigation algorithms. We also describe the sensor chosen for the MAV05 competition, held in Garmisch Partenkirchen, and a high level system and multi-modal interface providing more advanced autonomy in terms of collision free path planning and in-flight mission reconfiguration.

A Low-Level Active Vision Framework for Collaborative Unmanned Aircraft Systems

Micro unmanned aerial vehicles are becoming increasingly interesting for aiding and collaborating with human agents in myriads of applications, but in particular they are useful for monitoring inaccessible or dangerous areas. In order to interact with and monitor humans, these systems need robust and real-time computer vision subsystems that allow to detect and follow persons.

HDRC3 - A Distributed Hybrid Deliberative/Reactive Architecture for Unmanned Aircraft Systems

This chapter presents a distributed architecture for unmanned aircraft systems that provides full integration of both low autonomy and high autonomy. The architecture has been instantiated and used ...

Model-predictive control with stochastic collision avoidance using Bayesian policy optimization

Robots are increasingly expected to move out of the controlled environment of research labs and into populated streets and workplaces. Collision avoidance in such cluttered and dynamic environments is of increasing importance as robots gain more autonomy. However, efficient avoidance is fundamentally difficult since computing safe trajectories may require considering both dynamics and uncertainty. While heuristics are often used in practice, we take a holistic stochastic trajectory optimization perspective that merges both collision avoidance and control. We examine dynamic obstacles moving without prior coordination, like pedestrians or vehicles. We find that common stochastic simplifications lead to poor approximations when obstacle behavior is difficult to predict. We instead compute efficient approximations by drawing upon techniques from machine learning. We propose to combine policy search with model-predictive control. This allows us to use recent fast constrained model-predictive control solvers, while gaining the stochastic properties of policy-based methods. We exploit recent advances in Bayesian optimization to efficiently solve the resulting probabilistically-constrained policy optimization problems. Finally, we present a real-time implementation of an obstacle avoiding controller for a quadcopter. We demonstrate the results in simulation as well as with real flight experiments.

A Collaborative Framework for 3D Mapping Using Unmanned Aerial Vehicles

This paper describes an overview of a generic framework for collaboration among humans and multiple heterogeneous robotic systems based on the use of a formal characterization of delegation as a speech act. The system used contains a complex set of integrated software modules that include delegation managers for each platform, a task specification language for characterizing distributed tasks, a task planner, a multi-agent scan trajectory generation and region partitioning module, and a system infrastructure used to distributively instantiate any number of robotic systems and user interfaces in a collaborative team. The application focusses on 3D reconstruction in alpine environments intended to be used by alpine rescue teams. Two complex UAV systems used in the experiments are described. A fully autonomous collaborative mission executed in the Italian Alps using the framework is also described.