Road safety is a critical concern for all stakeholders in the automotive industry. While vehicle safety has come a long way, ensuring the safety of Vulnerable Road Users (VRUs) outside the vehicle remains a complex challenge. VRUs include pedestrians, cyclists, motorcyclists, and individuals with disabilities or reduced mobility and orientation. emotion3D, ZKW and TU Vienna have taken a step forward to address the issue of VRU detection with a holistic and innovative approach.
The world of camera-based analysis of automotive environments can be broadly divided into two areas: interior analysis and exterior analysis. Interior analysis primarily focuses on the safety of the driver and passengers within the vehicle. Features like seatbelts and airbag deployment systems aim to reduce injuries in case of an accident while driver monitoring systems help prevent the driver from falling into distraction or drowsiness states. On the other hand, exterior analysis concentrates on the safety of VRUs outside the vehicle, where detecting and avoiding potential collisions is of utmost importance. In addition, a detailed understanding of VRUs can result in more reliable and forward-thinking systems for driver assistance (ADAS) and automated driving (AD).
Detecting VRUs in the external environment is challenging for a multitude of reasons. One of the significant hurdles is poor visibility caused by adverse weather conditions or low light, such as during nighttime or fog. As the distance between the VRU and the vehicle increases, the accuracy of detection decreases, making it difficult to anticipate potential hazards in time. Additionally, certain positions, angles and movements of VRUs relative to the vehicle can lead to detection issues, especially in crossing situations. These factors collectively limit the effectiveness of VRU detection systems.
In this project, emotion3D extends its AI software stack and expertise for advanced human analysis to enable precise VRU analysis in the exterior of vehicles. The SmartProtect project, which is executed together with ZKW, a large supplier of premium-lighting systems and electronics, and TU Vienna, addresses the problem of VRU detection with a holistic approach combining imaging and distance-detecting sensor technologies. This novel automotive perception system uses automated, context-sensitive headlights and infrared lighting to ensure proper visibility of VRUs in images and LiDAR point clouds under all conditions.
The scope of the SmartProtect project is set to extend beyond theoretical development. emotion3D and its project partners extensively tested the envisaged multimodal systems in real driving conditions and representative, reproducible test scenarios. This rigorous evaluation process ensured that the solution is not only effective in controlled environments but also robust enough to perform reliably in real-world situations.
Road safety remains a top priority for the automotive industry, and addressing the detection of Vulnerable Road Users is an integral part of ensuring safer roads for everyone. The SmartProtect project stands as a testament to the company’s commitment to innovation and its vision of safer roads for all road users. Through ongoing research and development, emotion3D is paving the way for a future where road accidents involving VRUs become a thing of the past.
Disclaimer: The SmartProtect project has received funding from the Mobility of the Future programme. This programme is a research, technology and innovation programme of the Republic of Austria, Ministry of Climate Action and is managed by the Austrian Research Promotion Agency (FFG).