The loops sub-team is focused on building models to analyze whether to take loops during the race. The loops team is building models for before and during the race.

Benefit-Cost Score

Project Summary

This model scores loops prior to the race beginning based on various factors.

Goal

The goal of this model is to be able to calculate the “efficiency score” or reward for each loop to determine which one to take.

Inputs

For each loop we want to determine a cost to benefit ratio based on the loop properties, which include:

Distance

How much the additional distance will increase our score.

Stop/Start Moments

Accelerating/ Decelerating from/to stop will have an effect on energy consumption so we need to look at how we can minimize this. This isn’t an additional variable it directly impacts our power expenditure amount. Some extra reasons for stopping could be; Railroad tracks, weather, pedestrian crosses, or other stop causes not specified.

Speed Limits

The speed limit of the current road will also have an effect on the power that we consume. We need to look at how much power will be consumed by the current speed limit but also the km/h increase to a different speed limit.

Turns and their Steepness (90 degree or 120 degree)

Turns will also have an effect on our score. Not only using more power they also give us more points:

• 3 Levels 

  • 1-easy

  • 2-medium

  • 3-hard

• Multiply number of turns by difficulty of each

• Example:

  • 2 easy turns 

  • 3 medium

  • 1 hard

  • Total Turn Score: 2(1)+3(2)+ 1(3)= 11

Estimated Completion Time

We need to look at being able to estimate the total minutes for each loop. This can be calculated simply by using speed / distance, or as we take in more variables can get more accurate. The current status (20 miles behind projected distance or 15 miles ahead) will act as weight for this variable. We then want to be able to calculate the average loop completion time:

• If loop time > average loop, difference will act as negative and be affected by current status

• If loop time < average loop, difference will act as positive and be affected by current status

Output

...

Project Summary

Each driver will have different driving patterns when put behind the wheel of a solar car, coming in the form of different levels of battery efficiency, velocities, brake patterns, etc. In order for our models to more accurately predict battery efficiency, the driver must be taken into account and some adjustments may need to be made to the model depending on the person behind the wheel.

Goals

• Determine the types of data to profile our drivers (may be able to use inspiration from other teams driver profiles)

• Graph the data points to show trendlines for each driver

• Gain an understanding of areas that our drivers can improve or what our best driver is doing and how our other drivers can improve

Inputs

• Driver data points for velocity, time, battery efficiency levels, etc

Outputs

• Graphs will likely be velocity vs efficiency or something similar, something that will be able to display a driver’s efficiency based on time and speed

Notes

• This requires testing data to be obtained with the car and multiple drivers. Given our history, this probably won’t really be happening and may end up being used live at race during FSGP evaluating drivers over the course of race laps and analyzing the data on the fly