aikido icon indicating copy to clipboard operation
aikido copied to clipboard

Artificial Intelligence for Kinematics, Dynamics, and Optimization

Results 105 aikido issues
Sort by recently updated
recently updated
newest added

In preparation for #467, this PR contains the following update: - Add `planner::dart::ConfigurationToTSRwithTrajectoryConstraint` - Add `planner::dart::CRRTConfigurationToTSRwithTrajectoryConstraintPlanner` - Add `planner_dart` component and make relevant changes in CMakeLists files. The last change...

Currently, we almost exclusively use the CatkinResourceRetriever, which only accepts the `package://` schema. There are other ResourceRetrievers we can use that can accept other schemas (e.g. `file://`, `dart://`). DART has...

enhancement
good first issue

Currently, CatkinResourceRetriever builds up the list of packages from scratch using environment variables (e.g. CMAKE_PREFIX_PATH). This takes a while. But ROS already has a built-in way to resolve packages: https://wiki.ros.org/Packages...

enhancement
good first issue

XCode CI never successfully migrated to Github Actions. While this is low-priority (since all development is currently happening on Ubuntu), it should be fixed if possible.

enhancement
help wanted
good first issue

The collision constraint currently does not lock the skeleton before setting the state and evaluating the validity of the state. It looks like the hand frequently falls off when I...

The planning calls i.e. `Robot::PlanTo` and `Planner::Plan()` methods need to take in a termination condition.

There's an implementation for `MoveEndEffectorTwistVectorField` in the [`se2_policies`](https://github.com/personalrobotics/aikido/tree/se2_policies) branch, but we need to clean it up.

enhancement

Previously the skeleton was locked in `planTo...` methods before calling any of the planners to plan. In the new API this does not happen anywhere currently. I think it should...

bug

During the review of https://github.com/personalrobotics/aikido/pull/588, we decided it would be a good idea to move the VFP/CRRT params currently in `robot/util` into the corresponding planners (similar to what https://github.com/personalrobotics/aikido/pull/579 did...

enhancement

During the review of https://github.com/personalrobotics/aikido/pull/588, we realized that (for some reason) these methods take `StateSpace::State*` instead of `MetaSkeletonStateSpace::State*`. Since we're working with `ConcreteRobot` we must be working with `MetaSkeletonStateSpace::State*` (unless...

enhancement