Nichole L. Morris - US grants

This research project will develop a smart warning system to enable safe and minimally intrusive interactions between motorists and bicycles. Interactions with bicycles are rare for a typical motorist, therefore safety-conscious drivers naturally focus on other motor vehicles in the roadway, and may not become aware of the presence of a bicycle until it is too late. In contrast, interactions with motor vehicles are commonplace for a bicyclist. Furthermore, the bicyclist faces far greater consequences in an accident than a motorist. Therefore it is appropriate for a collision prevention system to be the responsibility of the cyclist. Continuous display of bright flashing lights or loud sounds may suffice to bring attention to the cyclist, but they may unnecessarily distract nearby motorists, or they may alarm passing drivers, and cause them to move dangerously far from their own lane. The system under development will guide motorists to pass bicycles with exactly as much distance as safety requires. Furthermore, it will provide alerts only to those drivers that have a significant probability of collision with the bicycle. The system to be developed will incorporate a knowledge base of likely collision scenarios, thus minimizing false alarms. The system will provide guidance cues to the bicyclist, to ensure a safe and respectful response to motor vehicles. Human factors studies will be used to design an alert system that provides motorists with specific and effective audio-visual cues. These studies will also be used to ensure that cyclists do not respond to the enhanced security by becoming more reckless. It is expected that the technology developed in this project will enable motorists to interact with bicycles safely and with minimal intrusion. It will reduce the approximately 48,000 bicyclist injuries and 700 fatalities that occur every year.

The development of a bicycle-mounted collision avoidance system must address a number of challenges beyond those required for a similar system on a car. These challenges include the need to address more complex collision scenarios, the need to provide alerts to the drivers of other vehicles, the need for inexpensive, light and smaller sensors, and the need to rely on human users for effective functioning of the system. These challenges will be addressed by development of unique custom-designed sensors, novel estimation algorithms for vehicle tracking and use of a rigorous human factors study to determine which warning systems will be effective and how such warnings should be provided to the involved motorist and bicyclist in real-world traffic scenarios. The warning presentation is designed to minimize the trade-offs between low reaction time and unnecessarily intrusive disturbances to nearby motorists. The custom sensors developed in the project include a triad sonar transducer unit for side vehicles, and front and rear laser sensors on real-time controlled rotational laser platforms to track vehicles at continuously changing lateral and longitudinal distances. The human factors studies in the project will enhance our understanding of human behavior in multi-modal collision avoidance systems and analyze possible long-term changes in behavior after prolonged use of the system. The project also includes an intensive 6-month field operational test in collaboration with an industrial partner to evaluate the effectiveness of the developed technology. The field operational tests will involve 10 bicycles, bicyclist volunteers with significant daily urban commutes and extensive analysis of bicycle data recorded in real-world traffic conditions. Due to the close industrial collaboration, the research conducted in this project will accelerate the path to commercialization of this smart system with its potential benefits to the country. The project will educate two graduate students and a post-doctoral researcher, providing them experience in inter-disciplinary research as well as an opportunity for strong industrial interaction.

This project is a collaboration between The University of Minnesota (Mechanical Engineering, Computer Science and Human Factors Engineering/Psychology) and primary industrial partner Quality Bicycle Products (QBP), (Bloomington, Minnesota, Large business). Broader context partners include The Minneapolis Bicycle Coalition, (Minneapolis, MN, nonprofit).

This Cyber-Physical Systems (CPS) grant will study smart tracking systems on scooters for ensuring safe and smooth interaction with other vehicles and pedestrians on the road. The smart system consists of inexpensive sensors, active sensing based estimation algorithms, and deep learning based robust image processing to enable trajectory tracking of all nearby vehicles on the road. If the danger of a scooter-vehicle collision is detected, an audio-visual alert is automatically provided to the car driver to make them aware of the presence of the scooter. The system also monitors the scooter rider?s behavior, provides real-time feedback to improve rider compliance with traffic signals and sidewalk rules, and documents the information as a part of the rider?s safety record. The key attractive features of the system are that it is inexpensive (< $500), is immediately useful on today?s roads without requiring the vehicles on the road to be equipped with additional technology, and is potentially commercializable. The project contributes to the society by improving safety of micro-transportation systems, and broadens participation in computing via undergraduate research activities and promoting significant cross-disciplinary collaboration between faculty in engineering, computer science and human factors.

The project will conduct research to develop two novel vehicle tracking technologies. The two technologies, one based on use of a low-cost single-beam laser sensor and another based on a low-cost low-density Lidar sensor, can have applications in protecting vulnerable transportation users such as bicyclists, motorcyclists, scooter riders, users in developing countries and also in other cyber-physical systems such as indoor robots. The computer vision system will handle rain, snow and low lighting which pose a major challenge by corrupting normal image data. New robust deep-learning-based recognition techniques will be developed that can effectively deal with corrupted image data sets. This will be achieved by novel nonlinear modeling of the low-complexity structures in both the clean data and in the image corruption using deep learning and allowing for mixed corruption types, varying severity and possible corruptions in the training data itself. To ensure human-in-the-loop robustness, the project utilizes human subject studies to evaluate the effectiveness of a variety of audio and visual mechanisms for alerting the motorist and the scooter rider, including innovations such as providing visual cues of biological motion on the scooter to improve localization of the scooter by motorists.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.