Performance evaluation of a radar-on-chip system for obstacle detection strategies for urban air mobility (UAM) applications

Purpose The transition to sustainable urban air mobility (UAM) is critical for decongesting ground traffic and reducing urban CO2 emissions through the use of electric aerial vehicles. However, this sustainability potential is contingent on solving the foundational challenge of safe, autonomous operations in dense urban airspace. This paper, therefore, analyses a “radar-on-chip” millimeter-wave Frequency Modulated Continuous Wave (FMCW) radar sensor (Texas Instruments IWR1642) as a core component for a detect-and-avoid (DAA) system, evaluating its performance as a key enabler for the safe and viable implementation of sustainable UAM. Design/methodology/approach Experimental results demonstrate that the IWR1642 is both precise and reliable in tracking moving objects, whether their motion is radial or transverse. Data processing was performed using the mmWave Demo Visualizer software, which utilizes the radar's basic detection algorithm. Using Geographical Positioning System (GPS) positions as reference, the radar's tracking of moving targets was found to be highly consistent, with only minor percentage errors during the data acquisition process. The radar was successively mounted onboard an unmanned aerial vehicle (UAV) to simulate autonomous operations in UAM scenarios, designing and developing a dedicated structure to house the sensor. The IWR1642 radar sensor demonstrated potential suitability for use in future UAM applications, providing accurate and reliable detection capability under a variety of conditions. Findings This work outlines the studies about possible solutions, identified by authors that could support beyond line of sight (BVLOS) operations in a medium- and low-risk environment, in particular, during UAM missions. UAM is considered a potential technology for future urban and suburban transportation, for both goods and passengers. By exploiting the vertical dimension, it decongests roads that are unable to support current city traffic while reducing CO2 emissions into the atmosphere, thereby moving toward a more eco-sustainable (through the use of exclusively electric vehicles) and multimodal transport system, in line with the concept of smart cities. This manuscript focuses on the operational limitations of the radar-on-chip system, as opposed to its functional constraints, emphasizing both the characterization phase and its performance during operational applications, such as in-flight testing. Based on the obtained results, the authors believe that the highlighted operational limitations could be mitigated by integrating additional sensors, such as light detection and ranging, to complement the radar-on-chip system. Originality/value This work represents a preliminary step toward the development of a robust obstacle detection system to be integrated onboard UAVs for UAM operations. UAS predominantly operate at very low altitudes (VLL, <500 feet), where the density of obstacles and human activity is significantly higher. Many of these aircraft function autonomously or in self-flying modes, necessitating rigorous safety measures to ensure mission reliability and operational safety. In this context, the ability to accurately perceive the surrounding environment is critical for UAM missions, particularly during BVLOS operations or fully autonomous missions. This study aims to develop a reliable obstacle detection system utilizing low-cost technologies.