Wireless Sensor Networks and Multi-UAV systems for natural disaster management

A method, apparatus, and computer program product for a drilling operation. In one advantageous embodiment, the drilling operation may comprise a motor, a rotary sensor, and a controller connected to the rotary sensor. The rotary sensor may be capable of generating a signal in response to rotary movement of the drill motor. The controller may be capable of monitoring a speed of the air motor from the signal generated by the rotary sensor and may be capable of identifying a change from a current layer in a stackup to a new layer in the stackup using the signal.

Adaptive positive feed drilling system

https://repositorio.inesctec.pt/bitstream/123456789/3974/1/P-00M-QZS.pdf

Survey on advances on terrain based navigation for autonomous underwater vehicles

Marine environments are delicate ecosystems which directly influence local climates, flora, fauna, and human activities. Their monitorization plays a key role in their preservation, which is most commonly done through the use of environmental sensing buoy networks. These devices transmit data by means of satellite communications or close-range base stations, which present several limitations and elevated infrastructure costs. Unmanned Aerial Vehicles (UAV) are another alternative for remote environmental monitoring which provide new types of data and ease of use. These aircraft are mainly used in video capture related applications, in its various light spectrums, and do not provide the same data as sensing buoys, nor can they be used for such extended periods of time. The aim of this research is to provide a flexible, easy to deploy and cost-effective Wireless Sensor Network (WSN) for monitoring marine environments. This proposal uses a UAV as a mobile data collector, low-power long-range communications and sensing buoys as part of a single WSN. A complete description of the design, development, and implementation of the various parts of this system is presented, as well as its validation in a real-world scenario.

Unmanned Aerial Vehicle Based Wireless Sensor Network for Marine-Coastal Environment Monitoring.

Effective simulation and testing environment is a vital part in the research of self-driving vehicles. It is capable of testing self-driving software and hardware quickly in a variety of virtual environments at low cost. However, as for the current mainstream simulation platforms, a considerable gap exists between the constructed virtual environment and the actual self-driving platform, which decreases the efficiency of development and makes it difficult to complete the migration from the virtual scenario to the real environment. Therefore, in this paper, we proposed a novel hardware-in-the-loop (HiL) simulation system. It takes the real hardware electronic control unit (ECU) of the self-driving vehicle as a part of the simulation platform, which improves the efficiency of development and testing, and also, the verified algorithms can be implemented into self-driving cars faster than ever before. The proposed HiL simulation system mainly consists of four parts: the vehicle kinematic model simulation, the multi-sensor simulation, the environment simulation, and the ECU hardware. Simulation experiments on applying the HiL system are used to verify the validity of self-driving algorithms in virtual scenes, including perception, planning, decision making, and control. Furthermore, algorithms that are tested in the simulation environment can be rapidly deployed into the real self-driving vehicles. In this paper, we also presented the verification processes of various algorithms, such as planning and control. These algorithms are implemented in the HiL system, and the experimental results show the validity of our proposed platform.

A Novel Integrated Simulation and Testing Platform for Self-Driving Cars With Hardware in the Loop

Beschrieben wird ein Optimierungsverfahren zur Regelung des Betriebszustandes einer gefuhrten Werkzeugmaschine mit einem, mit einer Drehzahl rotierenden und mit einer Schlagfrequenz Schlag-beaufschlagten Werkzeug wahrend eines Bohrvorganges, bei dem das Werkzeug zusatzlich kraftbeaufschlagt in einen, aus einem gegebenen Material bestehenden Gegenstand hineingetrieben wird. Die Erfindung zeichnet sich durch die Kombination der folgenden Verfahrensschritte aus: Erfassen von den Ist-Betriebszustand der Werkzeugmaschine charakterisierenden Parametern, Durchfuhren einer lernfahigen Prozessanalyse auf der Basis der erfassten Parameter zur Erhalt parameterspezifischen, relevanter Merkmale, Auswerten der parameterspezifischen, relevant Merkmale im Rahmen eines logischen Entscheidungsprozesses zum Erhalt einer den Ist-Betriebszustand charakterisierenden Information, Auswahl von bekannten, den Betriebszustand beschreibenden Referenzdaten in Abhangigkeit eines vorgebbaren Gutekriteriums, des eingesetzten Werkzeuges sowie des Materials des zu bearbeitenden Gegenstandes, Ermitteln eines durch Drehzahl und Schlagfrequenz bestimmbaren optimalen Arbeitspunktes fur die Werkzeugmaschine durch Vergleich der den Ist-Betriebszustand charakterisierenden Information mit den ausgewahlten Referenzdaten und Regelung der Werkzeugmaschine in Bezug auf Drehzahl und Schlagfrequenz auf der Grundlage des ermittelten optimalen Arbeitspunktes.

Optimierungsverfahren zur regelung des betriebszustandes einer geführten werkzeugmaschine mit einem rotierenden und schlag-beaufschlagten werkzeug während eines bohrvorganges

In order to reach optimal drill penetration using a rotary hammer, it is necessary to control the two variables, rotational speed and strike rate, of the rotary hammer in such a way that by applying minimal guidance and recoil power, a maximum drill penetration rate can be achieved in the rock. An optimal drill penetration rate is attained through using a given combination of rotational speed and strike rate. Changes in the drill diameter and/or material hardness lead to a false adjustment of the system which has to be re-optimized by an intelligent re-adjustment of the two servo-controlled drives. To achieve a flexible and automatic adaptation of rotational speed and strike rate of the rotary hammer to different material and tool types, an adaptive multi-sensor drive control based on a self-learning neuro-fuzzy component has been developed by IITB in cooperation with an industrial partner.

Smart neuro-fuzzy based control of a rotary hammer drill

In order to enable physical human-robot interaction where humans and (mobile) manipulators share their workspace and work together, robots have to be equipped with important capabilities to guarantee human safety. The robots have to recognize possible collisions with the human co-worker and react anticipatorily by adapting their motion to avert dangerous situations while they are executing their task. 
Therefore, methods have been developed that allow to monitor the workspace of mobile manipulators using multiple depth sensors to gather information about the robot environment. This encompasses both 3D information about obstacles in the close robot surroundings and the prediction of obstacle motions in the entire monitored space. Based on this information, a collision-free robot motion is planned and during the execution the robot continuously reacts to unforeseen dangerous situations by adapting its planned motion, slowing down or stopping. 
For the demonstration of a manufacturing scenario, the developed methods have been implemented on a prototypical mobile manipulator. The algorithms handle both robot platform and manipulator in a uniform manner so that an overall optimization of the path and of the collision avoidance behavior is possible. By integrating the monitoring, planning, and interaction control components, the task of grasping, placing and delivering objects to humans in a shared workspace is demonstrated.

Workspace monitoring and planning for safe mobile manipulation.

This contribution presents an approach to automatically classify crop plants and weed based on multi-sensor information with the aim of mechanically removing the weed by an agricultural robot. First, a possible sensor setup and its calibration are outlined. As the robot moves forward, the sensor data can be aggregated into a 3D model. To increase model quality, the roll and pitch angles of the robot have to be estimated and compensated. Then, different features are computed from the sensor data and the discriminative power of the features is evaluated. The feature vector is input to a support vector machine classifier. The considered classes are crop plant, weed, and soil. The result is a 3D representation of plants and weed which can be used for automatic weed removal. As an example application, tree nurseries are considered, especially the growth of boxwood trees. Classification and mapping results on real data acquired by a robot are reported.

/pdf/discrimination-of-plants-and-weed-by-multi-sensor-fusion-on-1uzwimo5ld.pdf

Discrimination of plants and weed by multi-sensor fusion on an agricultural robot

Deep learning (DL) is one of the key technologies in the artificial intelligence (AI) domain Deep learning neural networks (DLNN) profit a lot from the overall exponential data growth while on the other hand the computational effort for training and inference strongly increase. Most of the computational time in DLNN is consumed by the convolution step, which is based on a general matrix multiplication (GEMM). In order to accelerate the computational time for DLNN different highly optimized GEMM implementations for Graphic Processing Units (GPUs) have been presented in the last years [1] most of these approaches are GPU hardware specific implementations of the GEMM software kernel and do not incorporate the performance dependency of the training data layout. In order to achieve a maximum performance the parameters of the GEMM algorithm have to be tuned for the different GPU hardware and specific data layout of the training task. In this paper we present a two step autotuning approach for GPU based GEMM algorithms. In the first step the kernel parameter search space is pruned by several performance criteria and afterwards further processed by a modified Simulated Annealing in order to find the best kernel parameter combinations with respect to the GPU hardware and the task specific data layout. Our results were carried out on 160 different input problems with the proposed approach an average speedup against the state of the art implementation from NVIDIA (cuBLAS) from around 12 on a NVIDIA GTX 1080 Ti accelerator card can be achieved.

/pdf/gpu-gemm-kernel-autotuning-for-scalable-machine-learners-tv8ghwpbaj.pdf

Christian Frey

Papers

Optimierungsverfahren zur regelung des betriebszustandes einer geführten werkzeugmaschine mit einem rotierenden und schlag-beaufschlagten werkzeug während eines bohrvorganges

Smart neuro-fuzzy based control of a rotary hammer drill

Workspace monitoring and planning for safe mobile manipulation.

Discrimination of plants and weed by multi-sensor fusion on an agricultural robot

GPU GEMM-Kernel Autotuning for scalable machine learners