Fynn Schwiegelshohn

Computation and communication challenges to deploy robots in assisted living environments

Demographic and epidemiologic transitions have brought forward a new health care paradigm with the presence of both growing elderly population and chronic diseases. Recent technological advances can support elderly people in their domestic environment assuming that several ethical and clinical requirements can be met. This paper presents an architecture that is able to meet these requirements and investigates the technical challenges introduced by our approach.

An application scenario for dynamically reconfigurable FPGAs

Several recent studies have shown the technological feasibility of dynamically reconfigurable field programmable gate arrays (FPGA). In this paper, we introduce an application scenario from the area of autonomous robots which benefits from this technology as it uses elastic algorithms. The processing unit in an autonomous robot must handle localization, cognition, motion control, and perception. We analyze several elastic algorithms which solve the different tasks of the proposed application. Additionally, we suggest a design methodology for dynamically reconfigurable FPGAs that divides the problem into several steps and addresses them separately. In the first step, compiler optimization techniques are used to generate suitable data flow graphs. Then FPGA configurations are assigned to the tasks in these graphs. Finally, the configuration sequence with the minimal amount of reconfigurations is obtained by solving a Shortest-Common-Supersequence problem.

Enabling indoor object localization through Bluetooth beacons on the RADIO robot platform

Localization is one of the four pillars of the autonomous robotic control loop. In order to work in complete unknown indoor environments, the robot needs to map its surroundings. This is done via the simultaneous localization and mapping (SLAM) algorithm. However, the SLAM algorithm does not provide additional context to the generated map. If this information is required, it needs to be provided by the operator. With Bluetooth Low Energy (BLE) technology, position dependent information can be annotated to the generated map without operator input. BLE beacons need to be positioned at points of interest for the robot and then need to be localized. Because the BLE beacon broadcasts an ID, localization is based on the Received Signal Strength Indication (RSSI). This paper presents an approach to localize BLE beacons in the RADIO indoor environment. The robot has one BLE receiver which must be used cleverly in order to triangulate the BLE beacons position.

FPGA based traffic sign detection for automotive camera systems

Advanced driver assistance systems (ADAS) have become very prominent in todays automobiles. The technological advancement of already familiar assistance systems enable the car to now autonomously assess the current situation and react accordingly. In terms of data processing, there is no difference between actually acting i.e. accelerating, braking or steering and just issuing warnings to alert the driver of a dangerous situation. In this paper, we introduce a camera based image processing system for traffic sign detection for FullHD resolution. This system is able to detect speed limit traffic signs but additional traffic signs can be implemented using the same model. The hardware components consist of a Microblaze softcore from Xilinx and an extended IP core for HDMI-in and out signals. The system is implemented on a a Spartan-6-FPGA. For image acquisition, an off-the-shelf car camera is used. The developed system is able to reliably detect traffic signs on short distances on static images as well as on image streams.

Designing applications for heterogeneous many-core architectures with the FlexTiles Platform

The FlexTiles Platform has been developed within a Seventh Framework Programme project which is co-funded by the European Union with ten participants of five countries. It aims to create a self-adaptive heterogeneous many-core architecture which is able to dynamically manage load balancing, power consumption and faulty modules. Its focus is to make the architecture efficient and to keep programming effort low. Therefore, the concept contains a dedicated automated tool-flow for creating both the hardware and the software, a simulation platform that can execute the same binaries as the FPGA prototype and a virtualization layer to manage the final heterogeneous many-core architecture for run-time adaptability. With this approach software development productivity can be increased and thus, the time-to-market and development costs can be decreased. In this paper we present the FlexTiles Development Platform with a many-core architecture demonstration. The steps to implement, validate and integrate two use-cases are discussed.

Design of an attention detection system on the Zynq-7000 SoC

In this paper, we introduce a prototype attention detection system for automotive drivers. The driver is monitored through a Microsoft Kinect camera which provides RGB, depth, and infrared images in order to cover situations in which normal cameras might not achieve good results. The Kinect is connected to a Xilinx ZedBoard wich uses a Zynq-7000 SoC as processing platform. The attention detection system is running on the ARM Cortex-A9 dual core processor of the Zynq-7000 SoC. The system needs to recognize the drivers face and eyes in order to determine his state of attention. If the driver is classified as being attentive, no warning is generated. If the driver is classified as being inattentive, the system will generate a warning. Several algorithmic optimizations have been implemented in order to increase performance of this solution. In order to simulate a realistic driving environment, we have connected the Xilinx ZedBoard with a car simulator. This provides us with the necessary real world data to validate our system design. When our detection system classifies a driver as distracted or drowsy, it will send a warning message to the car simulator. The results show that the system performs satisfactorily when a face is detected. However, if no face is detected, the frame rate drops below an acceptable level.

A Fully Parallel Particle Filter Architecture for FPGAs

The particle filter is a nonparametric filter which approximates the posterior system state through a finite number of state samples i.e. particles drawn from a probability distribution. It consists of three steps which are motion update, sensor update and resampling. The first two steps are easily parallelized since the calculations do not depend on other particles. The resampling step however requires all particles to determine the particle set for the next iteration of the particle filter. In this paper, we introduce a novel FPGA optimized resampling (FO-resampling) approach to solve the parallelization problem of the resampling step by introducing virtual particles. Compared to multinomial resampling, FO-resampling achieves similar results with the added benefit of being able to completely parallelize all the steps of the particle filter. Additional to evaluating our approach with simulations, we implement a particle filter with FO-resampling on an FPGA.

Tackling The New Health-Care Paradigm Through Service Robotics: Unobtrusive, efficient, reliable, and modular solutions for assisted-living environments.

Medical care, supported by modern technology and methods, will play an important role in our global society. An increasing average age and the need to keep standards in care lead to the introduction of several solutions often described as ambient assisted living (AAL). However, most people feel uncomfortable with these solutions, describing a sense of secret observation. To alleviate this sense of secret observation, the sensor platform is placed in clear view of the end user. This is accomplished through the involvement of robotics in this domain, and both caregivers and the patient are offered support for several tasks. The Robots in Assisted Living Environments: Unobtrusive, Efficient, Reliable, and Modular Solutions for Independent Ageing (RADIO) project targets this topic but differs from other solutions by its specific concepts, which are especially the unobtrusiveness and inclusion into home automation. This article presents the RADIO approach with its solutions for combining robotics and home automation for use in AAL environments.

A resampling method for parallel particle filter architectures

Explanation of a new resampling scheme which uses virtual particles.Analysis of computational performance of the particle filter in general.Implementation of parallel particle filter with proposed resampling scheme. Particle filters are able to represent multi-modal beliefs but require a large number of particles in order to do so. The particle filter consists of three sequential steps: the sampling, the importance factor, and the resampling step. Each step processes every particle in oder to acquire the final state estimation. A high number of particles leads to a high processing time, thus reducing the particle filters usefulness for real-time embedded systems. Through parallelization, the processing time can be significantly reduced. However, the resampling step is not easily parallelizable since it requires the importance factor of each particle. In this work, a resampling scheme is proposed which uses virtual particles to solve the parallelization problem of the resampling component. Besides evaluating its performance against the multinomial resampling scheme, it is also implemented on a Xilinx Zynq-7000 FPGA.

Adaptive computing in real-time applications

Self-adaptive systems aim to satisfy system constraints in an environment with sudden parameter changes. Moreover, they enable new possibilities to cope with dynamic run-time requests. Their self-adapting capabilities lead to a highly dynamic system behavior which leads to development and runtime overheads. Additional real-time constraints increase the design space complexity further. The challenge is to reduce the complexity of application and system design. Recently, self-adaptive systems have been discovered for applications in emerging domains, such as the Internet of Things or Cyber-Physical Systems. Here, real-time support is often required for the respective systems. This paper surveys approaches to handle task scheduling/mapping, as well as resource management in these systems.