Road condition acquisition and assessment are the key to guarantee their permanent availability. In order maintain a country's whole road network, millions of high-resolution images have be analyzed annually. Currently, this requires cost time excessive manual labor. We aim automate process high degree by applying deep neural networks. Such networks need lot data trained successfully, which not publicly available at moment. paper, we present GAPs dataset, is first freely pavement distress...

Non-contact image photoplethysmography has gained a lot of attention during the last 5 years. Starting with work Verkruysse et al. [1], various methods for estimation human pulse rate from video sequences face under ambient illumination have been presented. Applied on mobile service robot aimed to motivate elderly users physical exercises, can be valuable information in order adapt conditions. For this paper, typical processing pipeline was implemented robot, and detailed comparison...

Analyzing scenes thoroughly is crucial for mobile robots acting in different environments. Semantic segmentation can enhance various subsequent tasks, such as (semantically assisted) person perception, (semantic) free space detection, mapping, and navigation. In this paper, we propose an efficient robust RGB-D approach that be optimized to a high degree using NVIDIA TensorRT and, thus, well suited common initial processing step complex system scene analysis on robots. We show superior RGB...

Humans are able to learn recognize new objects even from a few examples. In contrast, training deep-learning-based object detectors requires huge amounts of annotated data. To avoid the need acquire and annotate these data, few-shot detection (FSOD) aims instances categories in target domain. this survey, we provide an overview state art FSOD. We categorize approaches according their scheme architectural layout. For each type approach, describe general realization as well concepts improve...

The paper gives a comprehensive overview of our Shopping Guide project, which aims at the development interactive mobile shopping companion robots for everyday use in challenging operating environments such as home improvement stores. It is spanning an arc from expectations and requirements store owners customers, via challenges scenario environment, implemented functionality guide robots, up to results long-term field trials. trials started April 2008 still ongoing aim studying whether how...

This paper presents results of user evaluations with a socially assistive robot companion for older people suffering from mild cognitive impairment (MCI) and living (alone) at home. Within the European FP7 project "CompanionAble" (2008-2012) [1], we developed technologies combining mobile smart environment aim to support these assist them in their familiar home environment. For final evaluation, experience studies were conducted volunteer users who invited test where they lived freely used...

This paper presents the implementation and evaluation results of German research project SERROGA (2012 till mid 2015), which aimed at developing a robot companion for domestic health assistance older people that helps keeping them physically mentally fit to remain living independently in their own homes as long possible. The gives an overview developed robot, its system architecture, essential skills, behaviors, services required robotic assistant. Moreover, it new approach allowing...

The paper describes progress achieved in our long-term research project ShopBot, which aims at the development of an intelligent and interactive mobile shopping assistant for everyday use centers or home improvement stores. It is focusing on recent concerning two important methodological aspects: (i) on-line building maps operation area by means advanced Rao-Blackwellized SLAM approaches using both sonar-based gridmaps as well vision-based graph representations, (ii) a probabilistic approach...

Lidar based 3D object detection is inevitable for autonomous driving, because it directly links to environmental understanding and therefore builds the base prediction motion planning. The capacity of inferencing highly sparse data in real-time an ill-posed problem lots other application areas besides automated vehicles, e.g. augmented reality, personal robotics or industrial automation. We introduce Complex-YOLO, a state art network on point clouds only. In this work, we describe that...

This paper presents results of the development a socially assistive home robot companion for older people suffering from mild cognitive impairment (MCI) and living (alone) at home. work was part European FP7 project "CompanionAble" (2008-2012) [1] which aimed developing technologies that can support these elderly help them to remain in their familiar environment as long possible. To overcome current market entry barriers, start we consistently adopted user- application-centered process...

In this paper, we present MIRA, a new middleware for robotic applications. It is designed use in real-world applications and research teaching. comparison to many other existing middlewares, MIRA employs novel techniques communication that are described paper. Moreover, benchmarks analyze the performance of most commonly used middlewares ROS, Yarp, LCM, Player, Urbi, MOOS. Using these benchmarks, can show outperforms terms latency computation time.

Aging public roads need frequent inspections in order to guarantee their permanent availability. In many countries, this includes the standardized visual assessment of millions images. Due lack sophisticated approaches, often, evaluation is done manually and therefore requires excessive manual labor. GAPs most extensive publicly available dataset that provides standardized, high-quality images for training deep neural networks pavement distress detection. We further enlarge provide refined...

This paper presents the results of German research project SYMPARTNER (4/2015 - 6/2018), which aimed at developing a functional-emotional, mobile domestic robot companion for elderly people. The gives an overview developed robot, its system architecture, and essential skills behaviors required being friendly home companion. Based on this, in long-term field study running from January to June 2018 both technical aspects regarding practical suitability robustness under operating conditions...

Mobile service robots for transportation tasks are usually restricted to a barrier-free environment where they can navigate freely. To enable the use of such assistive in existing buildings, robot should be able overcome closed doors independently and operate elevators with interface designed humans while being polite passers-by. The integration these required capabilities an autonomous mobile is explained using example SCITOS G5 equipped differential drive Kinova Gen II arm 7 DoF. This also...

The paper gives an overview of the progress in developing a socially assistive home robot companion for elderly people with mild cognitive impairment (MCI) living alone at home. spectrum required functionalities such is broad and reaches from reminding functions (e.g. taking medication or drinking) stimulation exercises, via mobile videophony relatives caregivers, up to detection evaluation critical situations, like falls. addressing several aspects our work as part European FP7 project...

For robot mapping occupancy grid maps are the most common representation of environment. However, existing algorithms for creating such assume a fixed resolution cells. In this paper we present novel technique that chooses each cell adaptively by merging and splitting cells depending on measurements. The is based statistical measure derive in paper. contrast to other approaches adaption done online during process itself. Additionally, introduce Nd-Tree, generalization quadtrees octrees...

In this paper, we present a monocular, texture-based method for person detection and upper-body orientation classification. We build on commonly used approach recognition that uses Support Vector Machine (SVM) Histograms of Oriented Gradients (HOG) [1] but replace the SVM by decision tree with SVMs as binary makers. Thereby, in addition to pure persons, distinction eight classes is enabled. The humans estimation their from larger distances essential socially acceptable navigation mobile...

People detection in 2D laser range data is a popular cue for person tracking mobile robotics. Many approaches are designed to detect pairs of legs. These perform well many public environments. However, we working on an assistance robot stroke patients rehabilitation center, where most the people need walking aids. tools occlude or touch legs patients. Thereby, based pure leg fail. The essential contribution this paper generic distance-invariant scan features and distinction their With these...