Kensuke Harada's Webpage


Research on Grasp/Manipulation

Presentation materials on my recent results are available on SlideShare

Base Position Planning for Dual-arm Mobile Manipulators Performing a Sequence of Pick-and-place Tasks

A Manipulation Motion Planner for Dual-Arm Industrial Manipulators

For more information

Pick and Place Motion Planner for Dual-Arm Manipulators

ʐ^ 1. Pick-and-place motion planner composed of the offline and the online phases is proposed.
2. The offline phase calculates a set of regions on the object/environment surface.
3. The online phase plans the grasping posture, the object pose placed on the environment, the dual-arm manipulation strategy, and the trajectory between the initial and the final postures of the robot.
4. The effectiveness of the proposed method is confirmed by the experiment of the dual-arm robot HiroNX.

For more information

Object Placement Planner for Robotic Pick and Place Tasks

ʐ^ 1. The proposed planner automatically determines the pose of an object stably placed near a user assigned point on an environment surface.
2. The proposed method clusters the polygon model of both the environment and the object where each cluster is approximated by a planar region.
3. The object placement can be determined by selecting a pair of clusters from the object and the environment.
4. We impose several conditions to determine the pose of the object placed on the environment.

For more information

Probabilistic Approach for Object Bin Picking Approximated by Cylinders

ʐ^ 1. This paper proposes a method for object bin-picking for without assuming precise geometrical model of objects.
2. We consider the case where the shape of objects are not uniform but are similarly approximated by cylinders.
3. From the point cloud of a single object, we extract the probabilistic properties with respect to the difference between an object and a cylinder.
4. By using the probabilistic properties, the finger can maintain contact with the target object while avoiding contact with other objects.
5. To realize bin-picking, we approxime the region occupied by fingers by a rectangular parallelepiped.

For more information

Fast Grasp Planning for Hand/Arm Systems Based on Convex Model

ʐ^ 1. By the geometrical relashionship between convex models, we can easily define several grasping parameters.
2. Using the new approximation model and random sampling, calculation time can be reduced.
3. Several numerical examples are shown to confirm the effectiveness.

For more information

Research on Robotic Assembly

Snap Assembly System (Collaboration with Prof. Juan Rojas with SYSU, former postdoc of AIST)

ʐ^ 1. Constructed force controller based on the control basis approach for snap assembly
2. Assembly state identification based on force sensor information during snap assembly

For more information

Research on Humanoid Motion Generation

Whole Body Motion Planning for Biped Humanoid Robots

ʐ^ 1. Whole body motion planner combined with biped walking pattern generator
2. Motion Planning for Legged Robots on Varied Terrain
3. Simultaneous whole-body-motion/foot-step planner for biped humanoid robots

For more information