How to Handle Changing Constraints Dynamically in Acados?

[sehor]

Hello

I am working on an optimization problem using Acados where the constraints on the control variables need to change dynamically based on external conditions. Specifically; I want to modify upper and lower bounds on inputs in real-time without restarting the solver.

While I understand that constraints are typically defined during problem setup; I’m unsure of the best way to update them efficiently during execution.

Is there a built-in method in Acados to adjust constraints on the fly, or do I need to reinitialize the solver each time constraints change? Would modifying the acados_ocp structure & calling acados_update be sufficient / is there another recommended approach? Performance is a key concern; so I’d like to avoid unnecessary recomputations if possible.
Checked Python Interface — acados documentationTalend Course guide related to this and found it quite informative.

If anyone has implemented similar dynamic constraint handling in Acados; I’d appreciate insights into how you structured your solution. Are there specific functions / examples that demonstrate this capability? Any guidance on best practices would be really helpful!

Thank you !!

[FreyJo]

Hi,

you can update lbu and ubu for each stage individually, using the setter: constraints_set, see Python Interface — acados documentation

Best,
Jonathan