Sedat Dogru

Railway Fastener Inspection by Real-Time Machine Vision

In this paper, a real-time railway fastener detection system using a high-speed laser range finder camera is presented. First, an extensive analysis of various methods based on pixel-wise and histogram similarities are conducted on a specific railway route. Then, a fusing stage is introduced which combines least correlated approaches also considering the performance upgrade after fusing. Then, the resulting method is tested on a larger database collected from a different railway route. After observing repeated successes, the method is implemented on NI LabVIEW and run real-time with a high-speed 3-D camera placed under a railway carriage designed for railway quality inspection.

The 2017 Humanitarian Robotics and Automation Technology Challenge [Humanitarian Technology]

Presents information on the RASS 2015 Humanitarian Robotics and Automation Technology Challenge.

Energy Efficient Coverage Path Planning for Autonomous Mobile Robots on 3D Terrain

Coverage Path Planning (CPP) is an essential problem in many applications of robotics, including but not limited to autonomous demining and farming. Most works on CPP address time efficiency or coverage completeness in a bi-dimensional and flat environment, not taking the terrain relief into account. In this paper we use a Genetic Algorithm to optimize the solution to the CPP problem in terms of energy consumption, taking into account the constraints of natural terrains: obstacles and relief. Simulation results show that our approach is effective in reducing energy consumption of a mobile robot performing CPP.

Power Characterization of a Skid-Steered Mobile Field Robot

In this paper we study the power characterization of a skid-steered mobile robot operating both indoors and outdoors, taking the 3D elevation of the environment also into account. We show that the power equation of a driving DC motor together with the force model on an arbitrary terrain can be used to estimate the power consumption as well as the effective friction coefficients of a skid steered vehicle. We present a mathematical model that estimates friction due to skid as a function of slip angles, and the radius of curvature. We present experimental measurements and their statistical summaries regarding the power consumption of a skid steered robot operating at different radius of curvatures, which can be used for rough motion planning, as well as experimental results showing the fitness for our model.

Tracking a Sycophant Wireless Sensor Network for its seamless integration to mobile Wireless Sensor Networks

In this paper we introduce our novel concept of Sycophant Wireless Sensors (SWS) which is a static ecto-parasitic clandestine sensor network mounted incognito on a mobile agent using only the agents mobility without intervention. SWS networks not only communicate with each other through mobile WSN but also cooperate with them to form a global hybrid Wireless Sensor Network (WSN). Our work develops the SLAM methodology integrating the different type of mobilities of SWS networks and mobile WSNs where the units of the latter network need also to track SWS besides cooperating and maintaining connectivity between the different SWS networks to form a seamless hybrid sensor network. This paper focuses on that latter issue. Demonstrative results are provided on the experimental setup including performance analysis of the methodology which also is conducted in simulation besides the hardware implementation for an elaborate sensitivity analysis to system parameters.

Estimation of rotational speeds of skid-steered wheeled mobile robots using an improved kinematic model

It has been generally assumed that skid-steered wheeled robots have good odometry during translations, but unacceptable odometry during rotations. In this paper we propose an improved kinematic model for skid steered wheeled robots that enables directly using wheel rotations to estimate angular velocities and hence improve dead reckoning performance. The proposed model also takes into account position of the center of mass, giving good results even if the center of mass is displaced considerably.

Shape reconstruction using a mobile robot for demining and UXO classification

Metal detectors are widely used to detect and localize land mines, as well as metallic clutter that causes many false alarms. These false alarms are handled by manual inspection with a prodder or using extra features like size and depth as well as fusion with other sensors like ground penetrating radar or chemical sensors. Directly shape itself has never been used, mainly due to infeasibility of the results of the few existing studies, which require controlled environments and dense sampling of the target objects. In this paper, we propose a new method for shape reconstruction that works with sparse data, and show its feasibility using field data collected by a mobile robot equipped with a commercial pulse induction metal detector carried by a custom 2DoF arm. This paper also describes the method employed to sweep natural terrains with the mobile manipulator and provides results of imaging a set of different metallic objects.

The 2016 Humanitarian Robotics and Automation Technology Challenge [Competitions]

Presents information on the 2016 Humanitarian Robotics and Automation Technology Challenge.

Towards fully autonomous energy efficient Coverage Path Planning for autonomous mobile robots on 3D terrain

Coverage Path Planning (CPP) is an essential problem in many applications of robotics, including but not limited to autonomous de-mining and farming. Most works on CPP address time efficiency or coverage completeness in a bi-dimensional and flat environment, not taking the terrain relief into account. In this paper we use a Genetic Algorithm to optimize the solution to the CPP problem in terms of energy consumption, taking into account the constraints of natural terrains: obstacles and relief. Instead of requiring an elevation map of the environment, we also propose an autonomous sparse sampling of the environment which is used in conjunction with Kriging to successfully model the relief of the environment. Field tests confirm our energy consumption model for the robot, and simulation results show that our approach is effective in reducing energy consumption of a mobile robot performing CPP.

Power Characterization of a Skid-Steered Mobile Field Robot with an Application to Headland Turn Optimization

Skid-steered platforms are in use for many different purposes, including demining, military, construction and agriculture. Their power consumption varies considerably with the maneuver they are performing, depending heavily on its radius of curvature. Therefore, efficient operation of skid-steered platforms for any purpose requires proper path planning based on a mathematical model of their power consumption. With this fact in mind, this paper studies the power consumption characterization of skid-steered vehicles and presents a method based on physical principles to estimate friction on arbitrary surfaces, and then derives a mathematical model of friction in skid-steered platforms, showing that friction in such platforms depends on the radius of curvature and slip angles of the wheels. Afterwards, the derived model is used to show the optimum type of Π turns using a skid-steered platform in a coverage path planning scenario. The proposed friction model, as well as its forecast on the optimum Π turn, are verified using both indoor and outdoor field data.