2103483 Automotive System Design (2022/2)

Target

Our Goal

Our goal, as a first-time competitor with no former experience, is to finish in a good enough position to gain fame and recognition, so that in the next years to come we can have more resources and support to improve our capabilities and climb the leaderboard. We will determine the optimal finish that can set us apart from other competitors but is realistically achievable for a first-timer. Being a first-timer also means we have a lot of limitations which we will explain later.

Looking back at data from 2019, we can see that the performance level of the competition stays relatively unchanged. The best and worst time of the acceleration and skid pad event has changed for less than 0.3 seconds, so the time-to-score calculation will stay relatively unchanged. Thus, we will use the 2022 time-to-score calculation to set our target time.

performance Evolution

Reading the playing field

The competition is divided into two parts, static and dynamic events. To finish high on the leaderboard efforts must be put into both parts equally by the team, but our job and responsibility only extend to providing the team with the best car we can. Static events score is not directly reflective of the car's performance and thus can not be used to decide or dictate the direction of the car’s development. Thus, the first thing we did was re-ranking all the teams by considering only the dynamic events’ scores.

Dynamic score distribution

Then, we look at the total dynamic events’ score distribution. We can see that the grid is distributed like a negative logarithm graph with two outliers whose score is clearly ahead of other teams. But this graph is not completed since not every team participated in every event. This can happen due to many reasons such as reliability problem, team management problem, or simply the team’s decision to not participate, all of which is not related to the car's performance. So, we cannot judge the grid’s performance distribution by this graph alone. To do that we will fill the score sheet for every team by mathematical means.

First, we choose the acceleration and skid pad events as our primary focus, since both events are relatively simple to other dynamic events and have few factors other than the car performance itself. These two events will be important for deciding our standing later.

Since they are parts of the same event, the efficiency score is linked with the endurance score. And the endurance event is an elongated auto-cross event. If we can determine the autocross score from the last two events we can then fill in every event’s score. Luckily, the last two events are the acceleration and skid pad events which we deemed simplistic and reflective of the car’s performance.

Steps to determine our ideal finish

1. Score calculation

2. Score distribution analysis

3. Our target

Score calculation

First, we will see how the score of acceleration and skid pad is calculated. Plotting a simple graph of event time vs event score we can see that the score is given linearly proportional to the event time itself, except for some outliers whose time is so far off the rest of the competition. So we can assume that the score is proportional to the car performance itself.

Second, we plot the acceleration, skid pad, and autocross score to find a relation between the three. At first, there seemed to be many variations between them.

But after we look at the combined score of acceleration and skid pad events compared to the autocross score we can see some relation between the two.

A simple linear trend line is then used to predict the autocross score from the combined score of the acceleration and skid pad events.

Then the endurance score is simply 2.5 times the autocross score. Since they are the lap time of the same track only extended over multiple laps, we assume the car will perform the same in both events.

The efficiency score is calculated from an efficiency factor that came from a combination of total lap time and energy used. That means to achieve a high efficiency score, the team must finish fast while using little energy. Fortunately, using statistical linear regression between the endurance score, which came from the lap time, and the efficiency score produced an acceptable prediction of the score. This might be because unlike ICE powertrain efficiency does not vary much between teams.

Score distribution analysis

After we found the solution to predict every score from acceleration and skid pad score, we eliminate any team that didn’t participate in either, and formed a new ranking using the existing score for teams that have them, and fill in the blank score with our own calculation.

The resulting total dynamic score distribution shows a more evenly spread-out distribution with notable groupings. The leftmost group is the group of teams that have their acceleration and skid pad time so bad that they receive the same minimum score so they are tightly packed at the bottom of the leaderboard. The rightmost group is the top five and is comprised of teams that are well-funded and have a large number of members. That left us with the middle group which will be our target. In the middle of this group, there is a gap around the 300 points mark which also separate the top ten overall group. We will design a car with performance sufficient to finish in this spot. Now that we have our target score, we will look at the acceleration and skid pad score distribution to see if our goal is plausible.

Acceleration event

Looking at the acceleration score distribution, there are three peaks with the leftmost being the outliers whose time results in only the minimum points. From researching other teams, the middle peak is where the AWD team started to rank in the standing, and the rightmost peak is occupied by the top five, again finishing clear ahead of the grid. Since we are inexperienced with EV drivetrains, we decided that our design will be RWD so that it would be easier to design not only for the powertrain team but for the structure team and suspension team also, the reasoning will be discussed by the suspension team later. Knowing our limitations, we decide to finish at the bottom end of the middle peak, beating other RWD teams will be our target. That will give us 25 points and the 17th or 18th position in the acceleration event. To achieve that we will need to finish the acceleration event in 4.23 seconds according to the score calculation.

Even though we will not achieve the highest finish in the acceleration event, we don’t have any limitations for the skid pad event. So, we will try to achieve an above-average finish for this one. The distribution is more straightforward for this event as well. Apart from the leftmost outliers, the score is distributed almost linearly with a small gap on the right separating the top five again. Our goal, then, is to finish as “the best of the rest” which will net us 35 points and place us between 7th and 8th in the skidpad event by making a 4.9-second skid pad run.

Skid pad event

Final score prediction

Using the same linear trendline that we used to calculate every team’s score we predicted that we will get 316 points, netting us an eleventh-place finish. Adding two other trendlines predicting the minimum and maximum points we can get we calculated a score range of 265 – 360 points which will guarantee a minimum top- 15 finish.

Our Target

Acceleration event
- 25 points
- 4.23 seconds run
Skid pad event
- 35 points
- 4.9 seconds run
overall finish
- 300 points

PREVIOUS - Benchmarking

NEXT - Design Spec

Page updated

Google Sites

Report abuse