Search papers, labs, and topics across Lattice.
4
0
5
A single controller achieves 40%+ reduction in lateral tracking error and lap time in autonomous racing by fusing MPC with learned tire dynamics residuals.
Marrying iterative learning control with reinforcement learning yields safer and more stable DRL controllers for batch process control, addressing a key adoption barrier in industrial settings.
Autonomous racecars can now overtake rivals 51% faster and with 81% success by predicting their moves and planning dynamically feasible trajectories.
Autonomous racecars can now learn tire dynamics 71% faster and with 60% higher accuracy by "seeing" the road surface and remembering past driving behavior.
