Robot Motion and Sensor Measurements

Basic Motion

This interactive simulation shows three examples of a robot measuring its distance to a wall.

■ Stationary ■ Constant speed ■ P-controller

Noise amplitude: 0.2 m

Constant speed: 0.3 m/s

P-controller gain: 0.3 1/s

Target distance: 0.5 m

Try changing the noise amplitude, driving speed, and proportional controller gain. Observe how the distance measurements change over time.

Filters

RawMinimumMaximumAverage

Sensor measurements

Noise amplitude: 0.2 m

Outlier amplitude: 1 m

Ordered filtering

1. Minimum window: 3 samples

2. Maximum window: 3 samples

3. Average window: 10 samples

Sample interval: 0.1 s.
Window size 1 disables that filtering stage.

This graph shows the effect of different filters on a signal. Why does the maximum and minimum filters seem to "move the line" to the right?

Filters and P-Controller

■ Stationary ■ Constant speed ■ P-controller

Robot parameters

Noise amplitude: 0.2 m

Constant speed: 0.3 m/s

P-controller gain: 0.3 1/s

Target distance: 0.5 m

Moving average filter

Show filtered signals

Window size: 10 samples

Recent measurement weight: 0 %

Window duration: 1.0 s. Larger windows smooth the signal but increase lag.

This graph shows the effect of a filter and a P-controller together. How does changing the window size and weight impact the graph?

Telemetry

■ Ultrasonic■ Encoder■ Objective

Robot motion

P-controller gain: 0.3 1/s

Objective distance: 0.5 m

Sensor parameters

Ultrasonic noise: 0.2 m

Encoder drift: 10 %

This graph is a pretty accurate depiction of what would happen with sensor drift if you didn't ever perform sensor integration.