Imitation Learning of Avoidance Modes for Efficient Multimodal Model Predictive Control
Overview
SideNet-MPC replaces MM-MPC's high computational cost with imitation learning, achieving 93% expert success rate in 18.6 ms per control period, 48% faster than the expert's 35.5 ms on sixteen threads.
Generated from attributed reports · Updated 2 hours ago
Event evidence and corrections
0 attributed source owners. Ownership does not establish independent confirmation. Quantities are reported separately and are never added together.
Reported quantity · control period: 18.6 other · Basis not reported
“keeps this rule and substitutes the network proposal for the minimum-cost mode. Hence at most two OCPs are solved per control period, and every applied input still comes from an OCP with control barrier function constraints rather than from the network. In $100$ seed-matched closed-loop Monte Carlo runs of mixed traffic, SideNet-MPC reaches $93\%$ of the expert's success rate and needs \SI{18.6}{ms} per control peri”
Exact source · revision 1Source owner not reported
Report timeline
Follow attributed reports and material updates.
- arXiv Robotics — research abstractsImitation Learning of Avoidance Modes for Efficient Multimodal Model Predictive Control
This research proposes SideNet-MPC, which replaces the exponential computational cost of multimodal model predictive control (MM-MPC) with imitation learning. A graph neural network predicts the expert's minimum-cost mode, reducing the number of optimal control problems (OCPs) solved per control period. The method achieves 93% of the expert's success rate with 18.6 ms per control period, 48% faster than the expert's 35.5 ms on sixteen threads.
Timezone · UTC
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).