Browsing by Author "Wehrmeister, Marco A."
Now showing 1 - 5 of 5
Results Per Page
Sort Options
- Artificial intelligence architecture based on planar LIDAR scan data to detect energy pylon structures in a UAV autonomous detailed inspection processPublication . Ferraz, Matheus; Júnior, Luciano B.; Komori, Aroldo S.K.; Rech, Lucas C.; Schneider, Guilherme H.T.; Berger, Guido; Cantieri, Álvaro R.; Lima, José; Wehrmeister, Marco A.The technological advances in Unmanned Aerial Vehicles (UAV) related to energy power structure inspection are gaining visibility in the past decade, due to the advantages of this technique compared with traditional inspection methods. In the particular case of power pylon structure and components, autonomous UAV inspection architectures are able to increase the efficacy and security of these tasks. This kind of application presents technical challenges that must be faced to build real-world solutions, especially the precise positioning and path following for the UAV during a mission. This paper aims to evaluate a novel architecture applied to a power line pylon inspection process, based on the machine learning techniques to process and identify the signal obtained from a UAV-embedded planar Light Detection and Ranging - LiDAR sensor. A simulated environment built on the GAZEBO software presents a first evaluation of the architecture. The results show an positive detection accuracy level superior to 97% using the vertical scan data and 70% using the horizontal scan data. This accuracy level indicates that the proposed architecture is proper for the development of positioning algorithms based on the LiDAR scan data of a power pylon.
- A Comparison of Fiducial Markers Pose Estimation for UAVs Indoor Precision LandingPublication . Junior, Luciano Bonzatto; Berger, Guido; Oliveira Júnior, Alexandre de; Braun, João; Wehrmeister, Marco A.; Pinto, Milena F.; Lima, JoséCooperative robotics is exponentially gaining strength in scientific research, especially regarding the cooperation between ground mobile robots and Unmanned Aerial Vehicles (UAVs), where the remaining challenges are equipollent to its potential uses in different fields, such as agriculture and electrical tower inspections. Due to the complexity involved in the process, precision landing by UAVs onmoving robotic platforms for tasks such as battery hot-swapping is a major open research question. This work explores the feasibility and accuracy of different fiducial markers to aid in the precision landing process by a UAV on a mobile robotic platform. For this purpose, a TelloUAVwas used to acquire images at different positions, angles, and distances from ArUco, ARTag, and ArUco Board markers to evaluate their detection precision. The analyses demonstrate the highest reliability in the measurements performed through the ArUco marker. Future work will be devoted to using the ArUco marker to perform precision landing on a mobile robotic platform, considering the necessary adjustments to lessen the impact of errors intrinsic to detecting the fiducial marker during the landing procedure.
- Sensor architecture model for unmanned aerial vehicles dedicated to electrical tower inspectionsPublication . Berger, Guido; Braun, João; Oliveira Júnior, Alexandre de; Lima, José; Pereira, Ana I.; Valente, António; Soares, Salviano; Rech, Lucas C.; Cantieri, Álvaro R.; Wehrmeister, Marco A.This research proposes positioning obstacle detection sensors by multirotor unmanned aerial vehicles (UAVs) dedicated to detailed inspections in high voltage towers. Different obstacle detection sensors are analyzed to compose a multisensory architecture in a multirotor UAV. The representation of the beam pattern of the sensors is modeled in the CoppeliaSim simulator to analyze the sensors’ coverage and detection performance in simulation. A multirotor UAV is designed to carry the same sensor architecture modeled in the simulation. The aircraft is used to perform flights over a deactivated electrical tower, aiming to evaluate the detection performance of the sensory architecture embedded in the aircraft. The results obtained in the simulation were compared with those obtained in a real scenario of electrical inspections. The proposed method achieved its goals as a mechanism to early evaluate the detection capability of different previously characterized sensor architectures used in multirotor UAV for electrical inspections.
- Sensorial testbed for high-voltage tower inspection with UAVsPublication . Berger, Guido; Oliveira Júnior, Alexandre de; Braun, João; Lima, José; Pinto, Milena F.; Valente, António; Pereira, Ana I.; Cantieri, Álvaro R.; Wehrmeister, Marco A.This work presents a methodology for characterizing ultrasonic and LASER sensors aimed at detecting obstacles within the context of electrical inspections by multirotor Unmanned Aerial Vehicles (UAVs). A set of four ultrasonic and LASER sensor models is evaluated against eight target components, typically found in high-voltage towers. The results show that ultrasonic sensor arrays displaced 25 ∘ apart reduce the chances of problems related to crosstalk and angular uncertainty. Within the LASER sensor suite, solar exposure directly affects the detection behavior among lower power sensors. Based on the results obtained, a set of sensors capable of detecting multiple obstacles belonging to a high-voltage tower was identified. In this reasoning, it becomes possible to model sensor architectures for multirotor UAVs to detect multiple obstacles and advance in the state of the art in obstacle avoidance systems by UAVs in inspections of high-voltage towers.
- Using multi-UAV for rescue environment mapping: task planning optimization approachPublication . Rosa, Ricardo; Brito, Thadeu; Pereira, Ana I.; Lima, José; Wehrmeister, Marco A.Rescuing survivors in unknown environment can be extreme difficulty. The use of UAVs to map the environment and also to obtain remote information can benefit the rescue tasks. This paper proposes an organizational system for multi-UAVs to map indoor environments that have been affected by a natural disaster. The robot’s organization is focused on avoiding possible collisions between swarm’s members, and also to prevent searching in locations that have already discovered. This organizational approach is inspired by bees behavior. Thus, the multi- UAVs must search, in a collaborative way, in order to map the scenario in the shortest possible time and, consequently, to travel the shortest reasonable distance. Therefore, three strategies were evaluated in a simulation scenario created in the V-REP software. The results indicate the feasibility of the proposed approach and compare the three plans based on the number of locations discovered and the path taken by each UAV.