Robotic Grasp Selection for Unknown Objects in Clutter
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous robotic systems face challenges in efficiently picking and placing unknown objects from cluttered environments due to difficulties in discerning object boundaries and grasping strategies, leading to reduced throughput and potential damage to objects.
Innovation Solution
A robotic system equipped with multiple cameras for visual data processing, using point cloud information and image segmentation to determine object geometries and graspable features, and selecting appropriate grasping strategies based on scores for successful object manipulation without prior object identification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If autonomous robots are used to move objects continuously without rest, then productivity is improved, but the ability to handle unknown objects in cluttered environments deteriorates due to lack of human judgment
Solution Approach 1:
The robotic system performs self-learning through reinforcement learning, where the robot autonomously improves its object handling capabilities by learning from trial and error interactions with unknown objects in cluttered environments, eliminating the need for human programming or intervention
Solution Approach 2:
The system dynamically adjusts grasping parameters such as grasp force, gripper position, and approach angle based on real-time sensor feedback and learned policies, allowing the robot to adapt to diverse object properties including texture, shape, and fragility
2Device complexity
If traditional grasping tools are used on unknown objects, then device complexity is reduced, but the risk of object damage increases due to inappropriate grasping force or method
Solution Approach 1:
The grasping force and gripper configuration are dynamically adjusted based on real-time feedback from force sensors and vision systems, allowing the robot to apply appropriate force for each specific object while maintaining simple gripper hardware
Solution Approach 2:
The system uses force sensors, torque sensors, and vision systems to continuously monitor grasping conditions and provide feedback to the control algorithm, enabling real-time adjustment of grasping parameters to prevent object damage
3Measurement precision
If multiple cameras and sensors are used to identify unknown objects, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The robotic system employs a multi-functional sensor suite where cameras, depth sensors, and force sensors serve multiple purposes including object detection, geometry estimation, texture analysis, and grasping force control, reducing the need for specialized sensors for each function
Solution Approach 2:
The system replaces complex mechanical object identification systems with vision-based and sensor-based perception algorithms that use machine learning to infer object properties from visual and tactile data
Data Source
AI summary
A set of one or more potentially graspable features for one or more objects present in a workspace area are determined based on visual data received from a plurality of cameras. For each of at least a subset of the one or more potentially graspable features one or more corresponding grasp strategies are determined to grasp the feature with a robotic arm and end effector. A score associated with a probability of a successful grasp of a corresponding feature is determined with respect to each of a least a subset of said grasp strategies. A first feature of the one or more potentially graspable features is selected to be grasped using a selected grasp strategy based at least in part on a corresponding score associated with the selected grasp strategy with respect to the first feature. The robotic arm and the end effector are controlled to attempt to grasp the first feature using the selected grasp strategy.