Robotic Assembly of Arbitrary Parts Using Force-Torque Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional approaches fail to develop robust methods for autonomous assembly of complex-shaped parts with 6-dimensional uncertainty in tight-clearance environments, as they are limited by object modeling inaccuracies, pose sensing errors, and are not feasible for arbitrary-shaped multi-peg-in-hole tasks.
Innovation Solution
A 6-axis robot employs force and torque sensing to iteratively adjust placement trajectories, using a surface-based sphere tree representation and constrained optimization to estimate contact configurations, allowing for active recovery motions and successful assembly of complex-shaped parts by iteratively refining placement attempts based on sensed forces and torques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a robot follows a nominal assembly motion trajectory for tight-clearance assembly, then assembly precision is improved, but the system becomes highly sensitive to uncertainties in object modeling, pose sensing, and robotic motion
Solution Approach 1:
The system dynamically adapts the assembly trajectory by switching from a rigid nominal trajectory to a flexible probabilistic trajectory that can accommodate uncertainties. The robot continuously adjusts its motion based on real-time force/torque sensing and probabilistic reasoning about contact configurations, making the system both precise and robust to variations in object modeling and pose sensing.
Solution Approach 2:
The system employs force/torque sensing feedback during the assembly process to detect contact events and update the probabilistic model of the assembly state. This feedback loop allows the robot to distinguish between successful placement and blocked trajectories, and to adjust subsequent motions accordingly, thereby maintaining reliability despite uncertainties in the nominal trajectory.
2Reliability
If the robot uses force and torque sensing to detect blockage and iteratively adjust placement trajectories, then reliability is improved, but the complexity of the control system and computation increases
Solution Approach 1:
The system performs self-diagnosis by using its own force/torque sensing capabilities to detect contact configurations and determine whether a trajectory is blocked or successful. The probabilistic reasoning framework enables the system to self-correct by automatically adjusting trajectories based on sensed feedback, reducing the need for external intervention or complex pre-programming of all possible contingencies.
Solution Approach 2:
The system changes key parameters during assembly, including the probabilistic distribution of contact configurations, the estimated goal configuration, and the approach trajectory. By dynamically updating these parameters based on force/torque sensing, the system achieves high reliability without requiring a statically complex control architecture, as the complexity is managed through adaptive parameter adjustment rather than fixed complex structures.
3Productivity
If the robot attempts multiple successive alternate trajectories in response to blockage, then productivity is improved through recovery from failure, but the total time for assembly increases due to iterative adjustments
Solution Approach 1:
The system performs preliminary probabilistic reasoning about possible contact configurations and potential blockage points before executing the assembly trajectory. By pre-computing probable contact configurations and preparing alternate trajectories in advance, the robot can quickly switch to recovery motions when blockage is detected, minimizing the time lost during iterative adjustments while maintaining high productivity through successful assembly completion.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The method effectively handles 6-D uncertainties, achieving high success rates in complex multi-peg-in-hole assembly tasks with orientation and position clearances of less than 0.015 rad and 1.5 mm, respectively, by iteratively refining placement attempts and adapting to uncertainties through force and torque feedback.
Implementation Method 1
Blockage is determined by force and torque sensing encountered by the end effector prior to attaining an intended pose or position
Implementation Method 2
Blockage is determined by force and torque sensing encountered by the end effector
Implementation Method 3
The robotic member has an end-effector at a distal end configured for compressively grasping the object
Data Source
AI summary
A 6-axis robot performs autonomous complex assembly based on a task-independent approach by directing a robotic member on a first approach trajectory for placement of a grasped object, and determining that placement along the first trajectory is blocked. The robotic member has an end-effector at a distal end configured for compressively grasping the object. Blockage is determined by force and torque sensing encountered by the end effector prior to attaining an intended pose or position. The robot identifies a second, alternate approach trajectory for placement of the grasped object based on a probability of success of the second approach trajectory, resulting from considering probabilities of candidate trajectories or positions. The robotic member continues iterative placement based on successive, alternate trajectories until placement is achieved. The robot attempts a successive placement around localized positions/trajectories in response to increased probabilities and decreased distance/error from the desired pose.