Predictive Robot Handover Control for Smooth Human Interaction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robotic systems face challenges in performing smooth, intuitive, and reliable handover tasks with humans, often resulting in unexpected contact or camera obstructions due to jerky movements and inadequate path planning.
Innovation Solution
A reactive and predictive robot control system using model predictive control (MPC) optimizes robot motions over multiple time steps, incorporating constraints to ensure smooth and predictable handovers, while integrating learning-based grasp planners and force classifiers to enhance safety and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If the robot moves quickly to grasp the object, then the handover speed is improved, but the motion becomes jerky and may cause unexpected contact with the human
Solution Approach 1:
The system dynamically adjusts robot motion parameters by optimizing sequences of motions over multiple future time steps using model predictive control. The system continuously adapts the motion plan based on predicted human hand position and velocity, allowing the robot to move quickly when safe and slow down when proximity to the human hand increases, thereby achieving both speed and smoothness
Solution Approach 2:
The system performs preliminary planning by optimizing a sequence of future motions before executing them. By predicting the human hand's future position and velocity over multiple time steps, the robot can pre-calculate a smooth motion trajectory that avoids jerky movements while maintaining efficient handover speed
2Manufacturing precision
If the robot approaches the human hand closely to grasp the object, then the grasp accuracy is improved, but the likelihood of contact with the human increases
Solution Approach 1:
The system uses continuous feedback from force sensors and camera tracking to monitor the distance between the robot gripper and human hand. This feedback is fed into the model predictive control system, which adjusts the motion plan in real-time to maintain optimal grasp accuracy while preventing contact by reducing speed or adjusting trajectory when the human hand position changes
Solution Approach 2:
The robot dynamically adjusts its approach speed and trajectory based on real-time predictions of human hand motion. By continuously optimizing the motion sequence over multiple time steps, the system can approach closely enough for accurate grasping while maintaining enough flexibility to prevent contact if the human moves their hand unexpectedly
3Reliability
If the robot moves smoothly and predictably, then safety is improved, but the handover time increases
Solution Approach 1:
The system maintains continuous optimization of motion sequences over multiple future time steps, ensuring that the robot is always moving along the most efficient safe trajectory. By continuously updating the motion plan based on predicted human hand position, the robot maintains smooth and predictable motion without unnecessary pauses or delays, achieving both safety and efficiency
Solution Approach 2:
The robot performs preliminary optimization of motion sequences before execution, calculating the entire trajectory in advance based on current predictions. This allows the robot to move efficiently along a pre-planned smooth path without making reactive adjustments that would cause jerky movements or delays, thereby achieving both safety and fast handover
4Reliability
If the robot uses complex path planning to avoid contact, then safety is improved, but the computational complexity increases
Solution Approach 1:
The system uses dynamic model predictive control that optimizes motion sequences based on predicted human hand position and velocity. This dynamic approach allows the robot to handle complex safety constraints efficiently by continuously adapting the motion plan to current conditions, achieving high safety without requiring overly complex static path planning algorithms
Solution Approach 2:
The system changes the control parameters by optimizing sequences of motions over multiple future time steps rather than simple point-to-point control. This parameter expansion allows the robot to incorporate safety constraints and human motion predictions into the optimization, achieving improved safety through mathematically tractable parameter adjustments
Data Source
AI summary
Approaches presented herein provide for predictive control of a robot or automated assembly in performing a specific task. A task to be performed may depend on the location and orientation of the robot performing that task. A predictive control system can determine a state of a physical environment at each of a series of time steps, and can select an appropriate location and orientation at each of those time steps. At individual time steps, an optimization process can determine a sequence of future motions or accelerations to be taken that comply with one or more constraints on that motion. For example, at individual time steps, a respective action in the sequence may be performed, then another motion sequence predicted for a next time step, which can help drive robot motion based upon predicted future motion and allow for quick reactions.