People riding bikes in the Netherlands are very likely to be involved in accidents. They are vulnerable, easily overlooked and sometimes very distracted. In a world with A.I. based vehicles and sensor based road assistance in cars we should be able to help bike riders as wel.
That's why I designed the 'HOI safety detector' for bike riders. A device that helps bike riders to be more aware of their surroundings using sensors and by providing safety nudges.
My name is Cristina and I'm a graduating intern in Industrial Design at Mirabeau - a Cognizant Digital Business. My graduating project is part of Mirabeau's Innovation Practice. The goals of this practice is proactively broaden the technological and design horizon to better serve clients in a future.
Research shows: riding a bike is getting more hazardous
Before I could form my design and engineering thesis I needed to do some research. I started by doing desk research and interviews behind the cause of most accidents where bikes are involved.
Cycling injury is one of the most common accidents in the Netherland. Although the Dutch have an excellent road safety performance and are known for the extensive bicycle infrastructure, road safety for cyclists shows a less favourable development than other road users.
Riding a bike through a city like Amsterdam can be challenging due to the high congestion of cars. Though various sources we found that more than 30% of the bicycle accidents are collisions with cars.
My literature study on bicycle-to-car accidents showed that most accidents happen on a crossroad when:
- Both parties were driving straight‐on (40%)
- One party crossed the path of the other (28%)
- One party turned left while the second went straight‐on (12%)
- One party crossed the road laterally while the other went straight‐on (12%)
- One party turned left while the other was going straight‐on from the opposite direction (8%)
Showing that 75% of accidents, the contact point of impact was located on the car front, 20% on the side of the car and 5% on the rear of the car.
How to navigate roads without auditory cues due to headphones and electrical cars?
My thesis is that cyclists and pedestrians rely heavily on auditory cues (like sounds of an approaching car, people yelling and ringing their bicycle bell) to detect and localise fellow road users. Obviously car drivers need to use more visual cues.
I found two trends that negatively influence the ability of cyclists using sounds to create awareness of their surroundings.
The first trend is the rising number of quiet electric and hybrid cars on the road. The second trend is that more people, including bike riders, wear noise cancelling headphones, listen to (loud) music, or both.
For bike riders to cope with a decreasing amount of auditory cues we can forbid headphones and introduce motor sounds for electrical vehicles, but we can also assist them in other ways.
Enhance situational awareness through sensors
Together with my coaches at Mirabeau I theorised possible solutions to the problem. Our goal: "Make cyclists more aware of their surroundings and alert them of possible threads to their safety".
Through workshops and a conception phase we decided to develop the HOI safety detector. Like the name says, it detects and signals whenever there is traffic from a place where the bike riders cannot see. HOI might be perfect for people that cannot rely on auditory input.
To prevent accidents, I used the data of how most accidents happened and information like the cyclists reaction time in sudden situations. I also considered that the speed while cycling varies per cyclist and situation.
This all of course is also dependent on a person’s fitness, the road conditions but also the traffic and weather. Bike riders drive at an average speed of 20km/h. Our calculations show that with that speed the Brake Reaction Time is about 2 seconds, or translated to distance travelled: 11,112 meters.
So, HOI needs to detect upcoming traffic somewhere above that distance for bike riders to be able to react in time.
First technical prototype
The first technical prototype I made together with Pjotr. It was made to proof what technology is most suited to detect objects. After some research we came to the conclusion that 'lidar' was most suited for the range we had in mind.
LIDAR is a technology where the distance between sensor and an object is detected through a bundled light and a photosensitive sensor. At first we mounted the LIDAR-sensor on the front of the bicycle. By using a servomotor the lidar can move and 'scan' an 180 degree area. The information the lidar part of the prototype gathers makes it possible to detect objects from a fair distance.
My assignment is also focussed on packaging the project in a useable and appealing design. The design you see below was developed after several sketches, and studies of shapes.
I set a couple of design principles for myself to create a shape that resonates with the goals of the product.
- Human Friendly, emotionally-appealing and easy to fit into once life. I believe we spend a lot of time in front of flat screens and wanted to counter balance it. A human design answers the innate need for soft and tactile three-dimensional shapes.
- Optimistic and daring The design needs to be authentic, approachable, and humble.
- Colourful and fun I designed a colour palette to illustrate how the product can be both useful and stand out.
- Technological The simplicity of the form and function will help hide the interface. This might be something unexpected.
- Materials and shadows The combination of materials will allow users to understand which parts of the device are for input (receiving the data from the to be detected objects like cars) and which give an output (the led lights lighting up to advice you of a possible danger).
The Hoi Safety Detector should delight its users by having a brand-appropriate design and be a SMART device (S –Significant, M – Measurable, A – Attainable, R Relevant and T –Traceable).
The design process
After designing the shapes, I needed to fit the parts on the bike. I also designed clamping systems for each component. I measured the handle bar of different bikes. There are a lot of different handle bar designs, so I designed the a universal clamp that would fit most of the variations. Mirabeau offered one of their Mirabikes as testbed for my designs. I shaped my designs to work with this one.
I used the Ultimaker3 3D printer to create test-versions of the clamping system. Although this 3D printer can print with a high tolerance I found that making a 3D print that can actually clamp with the right amount of force and fit takes a few tries. Of course the design also needed to be able to contain the electronics like the lidar system, motor, batteries and lights. To make sure the entire package fitted correctly I created many variants.
Once the first prototype was done, it was time to test it with several users. In those tests I also got feedback that countered the studies that said most accidents involved a frontal collision with the bike. The feedback suggested that as people can't constantly look behind, it might be more interesting to use the sensor as a system to provide bike riders a rear-view sense. I decided to test this idea and see the pros and cons of it. Here is a diagram explaining how HOI would work at the back of the bicycle when the car is found at 12 meters range and with an angle of 45º.
Learning’s and next steps
Right now I'm finalising my shape designs and production designs. I'm also constantly testing the back of the bike design in terms of usability and aestethics. I will also do an extra study of alternative shapes, colours, textures and other specific characteristics to finalise the look of the product. So, stay posted for more content!
- PROCEEDINGS of the 3rd International Conference on Driver Distraction and Inattention, September 4-6, 2013, Gothenburg, Sweden (No. 15-P)
- Cycling and sounds: the impact of the use of electronic devices on cycling safety. (online). (Enquiry: 07/01/2019).