A Framework for Regression Testing of Outdoor Mobile Applications
0 likes119 views
The document proposes a capture and replay framework to automate regression testing of outdoor mobile applications using noisy sensor data. The framework captures real sensor data streams during outdoor use and replays them in the lab, simulating the application's execution. It also allows editing replay sequences and comparing new outputs to gold standards to evaluate quality metrics and detect regression errors in updated applications. An example use case is given for the PeakLens mountain peak identification app.
A Framework for Regression Testing of Outdoor Mobile Applications
- 1. A Framework for Regression Testing of Outdoor Mobile Applications Carlo Bernashina Roman Fedorov Darian Frajberg Piero Fraternali Department of Electronics, Information and Bioengineering, Politecnico di Milano
- 2. Introduction and motivation Outdoor mobile AR applications are popular in fields such as gaming, tourism and environment monitoring. They rely on the input of multiple and noisy sensors: • GPS • Compass • Accelerometer • Gyroscope • Camera The regression testing of such applications requires the reproduction of the real conditions in which the application works, which are hard to reproduce. We present a capture & replay framework that automates regression testing of mobile outdoor applications and computes quality metrics to trace regression errors.
- 3. Capture and Replay Testing Framework Capture : It executes in the mobile device and acquires the sensor data streams in real-time on the field from multiple sensors during the outdoor usage sessions. Replay: It permits the visual inspection of the app in lab conditions by simulating its execution taking in input the captured data streams. Besides, it produces an execution trace composed by the new computed outputs. Sequence editor: It permits the replay of sequences and manual correction of positioning errors, generating gold sequences for evaluation. Report: It enables testing new versions of the business logic of the app by assessing the defined quality metrics such as Real-Time Average Angular Error (RTAAE). It compares the output produced by the app business logic during replay steps with the reference outputs stored in the gold sequence.
- 4. Use case: PeakLens app PeakLens is an outdoor AR mobile application that identifies mountain peaks and overlays them in real-time on the view by using artificial intelligence.