Prediction Dashboard
Created on August 22|Last edited on August 22
Comment
Investigate the Teams Work
Who are We? 👀
Hey! We're the prediction team 🫡.
The goal of motion prediction is to predict an agent's future behavior in a dynamic and multi-agent environment. The agent here will refer to the traffic participants in real-world driving such as vehicles, pedestrians, cyclists, etc. Accurate motion prediction is vital for safe, comfortable, and efficient motion planning, another downstream AV task. Such examples of motion planning include:
- To know whether to yield to a vehicle if they are going to cut in front of our AV
- To decide the best time for our AV to merge into traffic
Motion prediction requires an understanding of both static and dynamic word context such as an observation of other agents.
Video
Add a comment