1 of 11

The Arizona STEM Acceleration Project

Developing Predictive Models

2 of 11

Developing Predictive Models

A 7th grade STEM lesson

Amanda Sibley

Scott Blevins

1/14/2024

3 of 11

Notes for teachers

At this point students have learned how to program their drones and have refined their programs through trial and error. Today they are going to be challenged to program a path through a maze without the opportunity to try the maze repeatedly. They will need to collect data to relate the power and time of their drones to distances they can record on their scale models.

Developing Predictive Models: Click here

Error Sources and Sensor Integration: Click here

Sensor Integration: Click here

Loops and Logic: Click here

Sensor Integrated Movement: Click here

Sensor Integrated Movement- Taking Flight: Click here

List of Materials

Charged drone
Charged controller
USB connection cable
Computer for programming
Tools to measure distance
Tools to measure angle of rotation

4 of 11

Standards

Arizona Science Standards

Science and Engineering Practices:

Develop a model

Develop a model that allows for the manipulation of a proposed object, tool, process or system

Using mathematics and computational thinking

Apply concepts of ratio, rate, percent, basic operations, and simple algebra to scientific and engineering questions and problems.

Arizona Mathematics Standards

7.RP.A Analyze proportional relationships and use them to solve mathematical problems and problems in real-world context.

5 of 11

Objectives:

I can develop a process of collecting data that allows me to define a predictive relationship between time and distance traveled by the drone.

6 of 11

Agenda (2 Days)

Day 1:

Students will be asked to develop a Blockly code that would fly their drone through the obstacle course pictured on the right. What parts of the program are they confident in and what parts are they not?

Students will devise a procedure for collecting data so they can more accurately predict the distance the drone will travel at 50% power for various amount of time.

Students will begin the process of collecting data and developing regression models of this data to be used as their predictive tools.

Agenda & Timeframe: Read (1 minute)

On day 1 students will simply need to first recognize that they currently don’t have a great way to predict the relationship between the power of the motors, the duration of the movement command, and the actual distance the drone will travel. This will motivate the need for data collection.

While students are working on their data collection procedures make sure to reinforce the importance of always starting their drone in the same conditions (direction and position) to have reliable data. They will need to assign roles to team members to ensure reliable reset of the drone, the accurate collection of distance data, and the reliable recording of the data.

Students will need to consider repeatability. They may need to run each test for time and distance several times to make sure to account for variability in the machine’s performance. That is always a variable, especially when dealing with a consumer priced product such as these drones.

7 of 11

Agenda (2 Days)

Day 2:

Students will use Google Sheets as a place to record their time and distance traveled data and will develop a regression to be model the relationship.

Based on these models students will develop a Blockly code, complete with power and time settings to navigate the obstacle course they were working with previously. They should reflect on how close these settings are to what they found through their process of trial and error. If they feel that they need to change their predictive model they should do so now.

With a confirmed predictive model have students develop a code that would get their drone through the provided course.

8 of 11

Intro/Driving Question/Opening

How can I relate the time of flight to the distance the drone will travel?

9 of 11

Hands-on Activity Instructions

Small groups of 2-3 students. Consider defining roles as follows:

Data recorder and code enabler
Drone reset
Distance measurement

Have students focus on just the power setting of 50%. Students will collect data relating time of movement and the distance the drone travels.
Students will use Google Sheets to analyze the collected data in a scatterplot and to develop a regression that can be used to predict this relationship.
Students will use the predictive model to develop a program to fly through a given obstacle course.

10 of 11

Assessment

Students will formatively assess their own work when applying their developed models to the drone flights from the previous day that worked.

Students will then use the predictive models to develop a code that should travel the drone through a new course on the first attempt.

11 of 11

Differentiation

Offer students different roles in the group that don’t require technical skills but would still allow for them to be engaged.

Offer more direct instruction to struggling students with the logistics of using Google Sheets as a regression tool.

Remediation

Extension/Enrichment

Remind students that the drone can fly with different power settings and that the ultimate goal is to navigate the course the fastest.

Have students consider how they can develop a model relating time and distance traveled at higher powers.