Search

Real time location systems (RTLSs) are becoming more relevant in a more data driven economy and society due to their wide range of application cases. When the location of an object needs to be tracked with high accuracy, ultra wideband (UWB) technology is usually the best option. Nevertheless, UWB ranging estimates are not completely immune to some sources of error such as non line of sight (NLOS) or multipath conditions. Thus, this paper proposes a real-time classification model based on machine learning (ML) to predict if received ranging estimates are in line of sight (LOS) or NLOS conditions and discard those in NLOS. However, it is also shown that classifying measurements as LOS or NLOS does not guarantee detecting inaccurate ranging estimates, since LOS measurements can also yield large errors. As an example, the ranging root mean square error (RMSE) of the data labelled as LOS in a UWB based localization system database in the literature is of 0.714 m, significantly higher than the theoretical accuracy of a UWB system. Thus, a novel ML-based classification model is proposed to predict the magnitude of the ranging error. After applying the proposed classification model in the same data, the ranging RMSE of those ranging samples classified as most accurate is of only 0.183 m, significantly lower than the best RMSE we can obtain on the classical LOS/NLOS classification approach.

Direct injection spark-ignition (DISI) engines have significant potential to improve vehicle fuel economy and CO2 emissions in the transportation sector. Recent years have seen rapid growth in the market penetration of DISI engines in light-duty vehicles as a result of the intrinsic benefits the technology offers. Nonetheless, DISI engines face important challenges regarding regulated emissions, especially in terms of particulate matter (PM). Soot particles represent significant health and environmental hazards; thus, current emission standards strictly limit PM emissions and future regulations will be even more stringent. The objective of this work was to provide novel insights, analysis, and diagnostic tools to study and mitigate some of the drawbacks of direct injection systems, without compromising the superior thermal efficiency of DISI engines. The studies focused on the behavior of fuel sprays, soot formation at cold start conditions, and spray impingement. The experimental approach leveraged high-speed imaging of fuel injection, combustion, and spray impingement in optically-accessible engines to better understand various in-cylinder phenomena. First, non-reactive spray imaging experiments were conducted at three engine speeds (300, 1000, and 2000 RPM) and two intake pressures (50 kPa and 98 kPa) to assess the effects of piston motion and fuel flash boiling on the behavior and time evolution of fuel sprays. The results showed the significant influence of the in-cylinder flow and plume collapse on spray development, liquid fuel distribution, and cycle-to-cycle variability. Second, different charge strategies at cold start conditions (20° C) were evaluated by imaging the injection and combustion processes, while measuring engine-out gaseous and soot emissions. The different fueling strategies considered up to two injection events spanning a range of injection timing form 300° to 25° before top death center (bTDC). Differences in combustion, soot sources, and PM characteristics particular to the cold-start environment were identified, specifically fuel-rich pockets and wetted surfaces. Results show that emissions and engine performance are dominated by the effects of the later injection. Finally, a novel diagnostic tool for fuel impingement was designed and demonstrated. A piston was instrumented with a prototype sensor and a telemetry technique was developed for non-contact data transmission. The system was tested in a static optical engine while high-speed imaging of the spray was simultaneously recorded. The sensor was used to measure films formed by different blends of ethanol and gasoline (E10 to E80) at varying injection pressures and injection durations. The response of the prototype sensor was fast enough to capture the sub-microsecond dynamics in the spray impingement event (< 0.5 ms). The results demonstrated the sensor has excellent resolution and sensitivity to detect spatial and temporal characteristics of the impingement and the films generated by a realistic fuel injection system. Overall, the outcomes of the present work provide better understanding of the relationships between fuel spray characteristics and engine performance, which directly inform and improve fueling strategies for this important class of internal combustion engines.