PRECOG: PREdiction Conditioned On Goals in Visual Multi-Agent Settings
This work addresses the challenge of predicting uncertain intentions of other drivers for autonomous vehicles, representing an incremental improvement with a novel conditional forecasting approach.
The paper tackles the problem of forecasting vehicle trajectories in multi-agent driving scenarios by introducing a probabilistic model that performs both standard and conditional forecasting based on the autonomous vehicle's goal, achieving substantially higher accuracy than existing state-of-the-art methods.
For autonomous vehicles (AVs) to behave appropriately on roads populated by human-driven vehicles, they must be able to reason about the uncertain intentions and decisions of other drivers from rich perceptual information. Towards these capabilities, we present a probabilistic forecasting model of future interactions between a variable number of agents. We perform both standard forecasting and the novel task of conditional forecasting, which reasons about how all agents will likely respond to the goal of a controlled agent (here, the AV). We train models on real and simulated data to forecast vehicle trajectories given past positions and LIDAR. Our evaluation shows that our model is substantially more accurate in multi-agent driving scenarios compared to existing state-of-the-art. Beyond its general ability to perform conditional forecasting queries, we show that our model's predictions of all agents improve when conditioned on knowledge of the AV's goal, further illustrating its capability to model agent interactions.