SkiWell
Stay safe and connected with handsfree accident detection
Group 16
Vibhuv Reddy '25
Mechanical Engineering
Elias Rosenberg '22
Computer Science
Justin Sapun '25
Computer Engineering, Quantitative Studies
Problem Statement
Skiers need a way to alert nearby skiers if they are stuck in a tree well, even if they’re injured, physically restrained, or unconscious.
Tree wells are deep hidden pockets of deep snow that form when low-hanging branches block snow from compacting around the base of the tree. Skiers or snowboarders can easily fall in head first and can be knocked unconscious. The loose snow makes movement and breathing difficult as any movement dislodges more snow. The current solutions for victims are to scream and wait for help from ski patrol or other skiers.
Magnitude and Frequency:
90% of those who fall into a tree well are unable to get out by themselves
20% of all ski-related fatalities are caused by tree wells
60 seconds is all it takes to suffocate for someone stuck in a tree well
Users/Purchasers
Users
Groups of skiers and snowboarders venturing in the backcountry
Purchasers:
Individuals or ski resorts/ski ways to be rented to customers
State of the Art
Handle Alarm
Alarm built into handle of ski pole
Can be pressed lightly to alert skiers ahead to get out of the way, or heavily to sound an alarm for help.
** Requires physical activation from user
Avalung Oxygen bag
Provides oxygen through a tube to user in case of emergency
** Uncomfortable: Skiers must keep the tube in their mouths most of the time
** Expensive: Air canister + bag = $300
** Black Diamond discontinued this product because of choking hazard concerns
AVR Systems
Accident Beacon
Sends a beacon to alert the police/ski patrol there’s been an accident
Uses sensors to track physical state
** Not available/sold to average skiers
** Doesn’t alert nearby skiers to come help you
Avalanche Beacon
In the case of an accident, non-injured skiers switch setting of beacon to receive and begin their search for other skiers
No physical activation from user required as standard setting of device is set to transmit
** Requires extensive knowledge, training and practice to operate the device effectively
** Expensive: $275
Specifications
We are primarily focused on building a product that should satisfy our design specifications for ease of use, safety, comfortable, and durability before cost. Available products for back country skiers are already in the hundreds, so we have a little wiggle room to use advanced technology to be able to ensure a reliable product is made from good quality parts.
Prototypes
Initial Prototype
Automatic accident detection using 3-dimensional accelerometer
** Could not detect stationary/slow accidents
Sound response system to alert nearby skiers
** Sound response was ineffective past 10 meters under layers of snow
Improvements
Automatic Detection
Using a pressure sensor, the microprocessor is able to recognize when the user is no longer holding the ski pole. *Threshold in place to allow for hand movement
Wireless Communication
The devices are able to communicate with each other through Bluetooth 2.4 GHz - 2.4835 GHz Band Range in the event of a crash to signal the other device
Distance Estimation
Upon detection of an accident, the device is able to send GPS coordinates to the paired receiver allowing approximate distance to be recognized by the user in the form of continuous audio.
Final Prototype
Power On/Off
With feasibility in mind, this button was designed to give the user an easy way to turn the device on and off with one hand.
Device Housing
Designed to have an opening hinge, latch, and rough fitment for the device assembly
Wireless Connection Demonstration
![](https://www.google.com/images/icons/product/drive-32.png)
GPS Distance Demonstration
![](https://www.google.com/images/icons/product/drive-32.png)
Testing
Bluetooth Range
0-60 meters – worked as expected
60-80 metres – worked with delay
80+ meters – did not work reliable
95+ meters - did not work
Distance could increase in more secluded areas due to the decreased amount of interference in the open channel.
Sensor Testing
Set an ice bath to stimulate a cold, wet environment. The test was conducted by submerging the sensor in the ice bath to see if there was a change in performance.
The test was conclusive that there was no change, signifying cold and wet conditions have no impact on crash detection.
Glove Test
Thick gloves (Gore Tex) did not prevent successful pressure readings based on our activation threshold of 40.
However, lower pressure sensing for general holding values. Note, the general holding value does not cause any problems, or prevent accident detection.
Without gloves: 250-300 avg
With gloves: 150-200 avg
Sound/Distance Analysis
Fabric simulates the sound absorption properties of sound.
When buried, sound alarm will not alert skiers further than 5 meters away.
More effective than screaming when mouth is covered.
When uncovered, sound alarm can be heard >100 meters away.
Battery Life
Li Rechargeable Battery (Benchmark)
AAA Alkaline Battery
We can see that there is an obvious difference between the two battery types. Most notably, the alkaline batteries are able to withstand cold temperatures by not dropping significant voltage for longer periods of time. The graph on the left was a benchmark test graph with a current load similar to our device. The graph on the right was a test conducted by the placing the device in the freezer and measuring the alkaline battery voltage at 60 minute intervals.
Response Time
Benchmark: Ski Patrol on average can take 5-20 minutes (or more) to respond to different kinds emergencies depending on ski locations (on-mountain, off-piste, backwoods).
3 Tests each for finding someone 80 meters away:
With Siren:
The average response time to find someone with the siren and the gps distance calculation is 7:36 minutes
Without Siren:
The average response time to find someone with only the gps distance calculation is 15:19 minutes.
User Feedback
Peter Reynolds:
Club Ski Member, Thayer Engineer
“This is actually a solid idea, and I could see it being sold to Skiways and rented, or bought by backcountry skiers if it was integrated into the handle or something…
Right now it’s just too bulky, but you could fix that with your own circuit, and it would be sick if you wrapped the power cylindrically around the ski pole. “
Michael Shiaras
Skier, ENGS21 Peer
“Despite the size of the box, it’s actually not that heavy, and holding it is still feels super normal. I’d be fine skiing with this, and it’ll only get more compact down the line.
Instead of making it an add-on to a ski pole, I think people would just buy a new pole with this built-in to be honest.“
Rahul Kolluri
Club Ski Member, Cross-country skier (at home)
“I’m a fan of the button system you guys have worked out here. It seems really easy to use, and I don’t have to think about it much…
I’m worried about false positives though. Like what if I forget to turn it off and it’s just going off for all the other receiving poles? What if I throw it and never go get it?”
Matthew Landry
Backcountry skier
“I wish I had this for backcountry skiing at home. When I would go out skiing with my little brothers I constantly had to keep an eye out for them and it was honestly really nerve-wracking. This seems like a nice layer of extra safety.
Comfort Analysis (9 user interviews)
Comfortable to use?
Sound response comfortable?
Ethics and Sustainability
Ethics
Accessibility:
Sound-reliant response system – May not accessible to hearing impaired.
Loud sounds – May not suitable for young children.
Exposure to sounds at 110 dB can increase risk of permanent hearing loss.
Disturbance:
Sound response can disturb/cause unnecessary panic with nearby skiers in the case of a false positive.
The risk of false positive must be emphasized to user. User will be strongly encouraged to wear ski pole straps.
Sustainability
Recyclable Plastics:
Case made from thermocouple polyurethane
Removable Case:
TPU case and electronics can be recycled separately
Sustainable Packing:
Packaged in corrugated cardboard boxes
Injection Molding:
Cuts down energy costs required to 3D print on a medium scale