(This project was financed by the Piedmont Region (POR FESR 2007/2013), the European Fund for Regional Development and the Republic of Italy.)
The Autumn concept changes the way we think about safety, and about how we interact with the car and the environment while driving. Autumn was a research project to identify innovative human-computer interaction solutions for driving safety. Envisioning new uses for the latest in sensor and communication technologies, the Autumn concept identified ways that increased information flow between surroundings, vehicle and driver at strategic moments can enhance the driver’s awareness of danger and actively prevent accident situations from arising.
Experientia developed a model of cognition, describing the driving process and where failures occur that lead to accident situations. The research team began by exploring existing models, but then developed a broader understanding of concepts such as “human error”, and the chain of events that leads to accidents. The result was the “Situated model of driving”, which postulates that driving is a system made up of continuous interaction between the driver, the vehicle and the environment, which occurs at a subconscious level in normal situations, and at a heightened level of conscious alertness during unexpected events.
The team then developed the Autumn concept, a system to help prevent accidents from occurring, through a series of advanced feedback mechanisms that aid the driver’s awareness of surroundings and potential dangers before corrective actions need to take place. The concept has been showcased in a first-person perspective video prototype. It was also presented at the Epic conference in Boulder, Colorado, 2011.
Road safety is a major societal concern. Most safety developments in the last 15 years have focused on passive vehicle safety, making the car more resilient against accidents. Instead, the Autumn project focused on preventive safety (prevention of accidents) and active safety (recovery in an emergency situation). The research began by asking how can we:
- assist drivers in the driving process from the beginning to end of a route?
- enhance drivers’ perception of their surroundings?
- support drivers to detect potential dangers?
- prevent shock due to an unexpected event?
Developing a cognitive model
The research phase began by exploring the existing models to describe accidents and why they happen. The most important of these was the TRACE (TRaffic Accident Causation in Europe) project’s linear model of how an accident happens, which saw accidents as a chain of events, where the accident was the result of the last action performed in the sequence. Experientia believed that the cause and prevention of accidents could be seen in a less linear way, and instead developed a cognitive model (the “Situated model of driving”) in which drivers, vehicles and the environment are viewed as elements that constantly influence each other. When these three elements are correctly interacting, driving occurs smoothly. When they are incorrectly interacting, accidents result. This focused the design phase on ways in which technology can improve the interaction among the elements, making the system more resistant to failures.
While most current safety automation solutions support correction in critical conditions, our approach focuses on increasing people’s sensing and detecting capabilities. Existing advanced car safety systems focus on reaction time, in the period from an unexpected event occurring until the accident. The car continuously senses the environment, but only provides information, warnings or takes control in the emergency phase, often in a very obtrusive way. However, the situated model shows that perception and action inform and affect each other continuously, even when the action is performed in a habitual, subconscious way. Therefore, systems should enhance the natural human ability to sense and correct.
Our concept is based on natural, continuous and non-obtrusive interaction.
The windshield occupies an area dedicated to primary driving tasks, such as controlling the car. Head-up display (HUD) technologies can transform the windshield into an area for displaying information. Products currently in the market merely replicate dashboard information. However, the location is ideal for information about driving activities, rather than entertainment or secondary tasks.
The area of visual comfort, beneath the horizon line, displays critical information, while the upper part of the windscreen displays secondary information, such as current conditions. The driver-side window shows information on passing vehicles.
The metaphor of shadow is used to represent the safety zone of the car. When the shadow overlays a specific external element, it helps identify dangerous situations. By making the invisible visible, and drawing the driver’s attention to important elements in the environment s/he may have missed, we enhance the driver’s sensory capacity.
Straight cruise helps drivers to verify their position on the road, and be more aware of drifting out of their lane.
Proximity monitoring visualises the distance between the driver and other vehicles on the road.
Side eye enhances the peripheral vision of the driver, visualising the movement of surrounding cars, particularly those in the driver’s blind spot.
Moving objects imagines a future use for the technology in which moving objects around the vehicle, such as people or bikes, are identified, highlighted and monitored, in situations with poor visibility.
The practical result of the research was the creation of a first-person perspective video prototype to illustrate the concept in action. Dynamic safety information is shown on a head-up display. The concept aims to demonstrate that a driver trained by continuous and non-obtrusive information recognises dangerous situations earlier and has a faster reaction time, improving safety. It operates at a visceral cognitive processing level.
Video prototype of concept