ProRRT is our own sampling-based motion planner, based on the Rapidly-exploring Random Tree (RRT) algorithm. It is designed to be fast, efficient, and easy to use. More specifically it implements a greedy version of the RRT-Connect algorithm. Its main features are:
- Fast: ProRRT is a highly optimized implementation of an RRT-Connect algorithm, with a focus on speed and search efficiency. It implements Halton sequences for a more efficient and deterministic sampling strategy.
- Deterministic: Unlike other sampling-based solvers, ProRRT is deterministic, i.e. it will always return the same solution given the same inputs, which enables reliable and predictable behavior. This is done by using a deterministic sequence of random numbers.
- Support for constraints: Off-the-shelf support for joint range constraints and Cartesian orientation constraints.
- Support for padding: ProRRT supports padding around the robot and obstacles, which can be useful to include an extra safety margin in the planning process.
- Control over trajectory parameters: ProRRT allows the user to control some of the output trajectory parameters, such as the maximum joint velocity and acceleration.
Relevant headers
1.9.8