Learning Agile Paths from Optimal Control
This work addresses motion planning for agile robots, offering a hybrid approach that is incremental in combining existing methods.
The paper tackles the problem of efficient motion planning for agile robots by training a machine learning model on optimal control outputs, achieving a solution that balances optimality and computational efficiency.
Efficient motion planning algorithms are of central importance for deploying robots in the real world. Unfortunately, these algorithms often drastically reduce the dimensionality of the problem for the sake of feasibility, thereby foregoing optimal solutions. This limitation is most readily observed in agile robots, where the solution space can have multiple additional dimensions. Optimal control approaches partially solve this problem by finding optimal solutions without sacrificing the complexity of the environment, but do not meet the efficiency demands of real-world applications. This work proposes an approach to resolve these issues simultaneously by training a machine learning model on the outputs of an optimal control approach.