In for instance a bin-picking scenario a robotic arm is tasked with picking up and moving objects. This might entail generating valid grasps based on sensor inputs, e.g., a depth camera, and finding a valid path for the robot to follow in order to pick up and move the object, while avoiding obstacles and taking other constraints into account. We offer expertise on development and testing of such path planning methods for both robot manipulator arms and mobile robot platforms. We hold competency on building easily deployable solutions based on standard tools such as ROS and MoveIt, as well as researching planning methods built to fit custom use cases and demands.
In many cases, however, objects in the environment around the robot or the robot base itself might be in motion. In such cases efficient real-time motion planning methods must be developed in order to take into account the changing environment. Through our research we hold expertise on optimization-based motion planners that try to plan motions that avoids or interacts with dynamic objects in real-time.