Motion Planning of Robot

How can we ensure the safe movement of autonomous systems in a multi-agent environment?"

Navigate robot in multi-agent environment. Ensuring safe, i.e., collision-free, motion for autonomous agents is an ongoing and challenging task. We proposed a safety-critical control framework for an ego agent moving among other agents. The approach infers the dynamics of the other agents online and incorporates the inferred quantities into the design of control barrier function (CBF)-based controllers for the ego agent. The inference method combines offline and online learning with radial basis function neural networks (RBFNNs). Additionally, we employ conformal prediction to quantify the estimation error of the RBFNNs for the other agents’ dynamics, generating prediction intervals to cover the true value with a user-defined confidence. Finally, we formulate a quadratic program (QP)-based controller for the ego agent to guarantee safety with the desired confidence level by accounting for the prediction intervals of other agents’ dynamics in the sampled-data CBF conditions.

CBF-QP with offline-online neural network inference.

References

2025

  1. L-CSS
    Conformal Prediction in the Loop: Risk-Aware Control Barrier Functions for Stochastic Systems with Data-Driven State Estimators
    Junhui Zhang, Bardh Hoxha, Georgios Fainekos, and 1 more author
    IEEE Control Systems Letters, 2025
  2. ACC
    Safety-Critical Control with Offline-Online Neural Network Inference and Adaptive Conformal Prediction
    Junhui Zhang, Sze Zheng Yong, and Dimitra Panagou
    In 2025 American Control Conference (ACC), 2025