Autonomous

Multisensor

Systems

The amount of sensor data provided by battery-driven, widely distributed devices is steadily increasing. Since sensor data are typically fed into information processing units, it is worth considering how information processing itself can be exploited to reduce communication and energy demands. For this purpose, this project focuses on information-processing techniques that can incorporate implicit information conveyed by the transmission mechanism. Although a sensor node decides not to send its data, the receiver can still leverage the absence of data to update its state estimates. For instance, sensor readings can be compared against a threshold to decide for a transmission. The receiver can translate this decision rule into information about the data although no transmission took place. Sender and receiver can negotiate such decision rules in order to minimize communication costs, on the transmitting end, and to maximize the retrievable information, on the receiving end. Since threshold-based strategies are far too restrictive for time-varying systems being observed, model-based and data-driven policies will be investigated.

This project primarily investigates stochastic decision rules to trigger transmissions. In contrast to deterministic triggers, stochastic mechanisms can preserve the Gaussianity of the implicit information simplifying the estimator design at the receiver. For instance, a Kalman filter only requires minor adaptions to incorporate implicit information when no transmission event is triggered. The goal of this project is to push the principles of stochastic triggering forward to establish a comprehensive framework of lazy estimation. First, the investigations are concerned with general properties and the design of intelligent trigger decisions to improve the effectiveness and robustness of lazy state estimation. These include model-based and data-driven trigger mechanisms, aperiodic and asynchronous transmission and processing times, as well as the study of unreliable communication links. The results provide the foundations for large-scale lazy estimation with respect to both multisensor systems and high-dimensional state representations. For instance, multiple systems collaboratively monitor a dynamic system and fuse exchanged sensor data and estimates. Such distributed data fusion problems lead to dependent trigger decisions that require self-adapting trigger mechanisms. In particular, the project considers applications in object tracking to evaluate the derived concepts. Lazy estimation shows great potential in the processing of neuromorphic sensor data.

View project in the research portal

As part of the DatAmount research project, methods are being developed that make it possible to create energy models of machine tools. These models are suitable for predicting the energetic behavior of machines for new products on the basis of small amounts of data. Since small series are often produced, especially in the SME context, in many cases, there is not enough data to train AI models. Physical modeling, on the other hand, is often very costly. Due to the required CO2 proofs and the climate targets set, companies thus find themselves in a field of tension. On the one hand, accurate models to predict the energy consumption of machines are necessary to remain competitive. On the other hand, the creation of such models is currently either very expensive or not possible. The current mostly manual prediction of energy consumption is also time-consuming and also person-bound. The approach presented here combines physical models of the energy behavior of machines with data-based machine learning models, with particularly data-efficient machine learning models being investigated. This enables an automated, accurate prediction of the energy consumption of machine tools. The benefit for SMEs is the efficient creation of models that can predict the energy consumption and CO2 emissions of new products. These predictions are often necessary to be considered in a tender as proof of energy and resource efficiency is often mandatory in bids from larger companies with CO2 reduction targets.

View project in the research portal

Problem Definition
Autonomously operating mobility systems or delivery services open up considerable development potential with regard to quality of life and services of general interest in non-urban areas such as in the former brown coal regions of Germany. However, assessing the potential success of micromobiles operating autonomously on footpaths and cycle paths requires comprehensive environmental information from the areas of operation, such as minimum path widths, the volume of foot traffic or sight lines. In particular, outside of large cities, this information is incomplete and heterogeneously structured.

Project Objectives
The aim of the project is to design strategies for the bicycle-based collection of environmental data that are relevant for the successful operation of autonomous micromobiles on footpaths (visibility of certain areas, infrastructure parameters, passenger volume, network coverage, environmental data). For this purpose, the project evaluates different collection methods with regard to the efficiency and quality of the aggregated information. The usability of the data will be examined in two concrete smart city/town application scenarios (rental bicycles with autonomous delivery mode and delivery robots) with corresponding studies. In this way, the project contributes to the data-driven development of smart mobility and logistics concepts that cover the specific features of different settlement areas.

View project in the research portal