Trajectory Planning for Collaborative Robots
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional trajectory planning systems for collaborative robots (Cobots) in human-robot collaboration settings are impractical due to long compute times, making them unsuitable for dynamic and cluttered environments where quick collision avoidance and efficient task execution are necessary.
Innovation Solution
A computationally lightweight trajectory generation method that determines optimal trajectories considering velocity, acceleration, jerk, and dynamic constraints, allowing for fast and scalable motion planning in cluttered environments, and enables learning of motion primitives from demonstrations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional trajectory planning systems are used for collaborative robots, then collision-free paths can be generated, but the compute time is too long (60 seconds for a 9-second trajectory)
Solution Approach 1:
The trajectory planning problem is segmented into multiple sub-problems using a hierarchical approach. The configuration space is divided into free space and obstacle space, and the planning is performed in phases: roadmaps are pre-computed for free space, then collision checking and local optimization are performed separately. This segmentation allows parallel computation and reduces the overall compute time while maintaining collision-free path generation.
Solution Approach 2:
Roadmaps are pre-computed in advance for the free space of the configuration space, storing valid paths and transitions before actual trajectory planning is needed. This preliminary action allows the system to quickly query pre-computed roadmaps during real-time operation without performing full collision checking, significantly reducing compute time while ensuring collision-free paths.
2Adaptability or versatility
If trajectory planning is performed continuously for dynamic environments, then adaptability to human operators and environment changes is improved, but computational load increases
Solution Approach 1:
The system implements dynamic trajectory planning where roadmaps are pre-computed for the static free space, but collision checking and path selection are performed dynamically based on real-time sensor data and environment changes. This allows the cobot to adapt continuously to moving human operators and obstacles while keeping computational energy consumption low by only computing what is necessary at each moment.
Solution Approach 2:
Instead of recomputing entire trajectories in response to environment changes, the system performs local collision checking and optimizes only the affected portions of the path. When obstacles or human operators move, the system updates local roadmaps or adjusts existing paths in the affected regions rather than recalculating globally, reducing computational energy while maintaining adaptability.
3Speed
If fast trajectory computation is implemented, then real-time responsiveness is improved, but trajectory optimization quality may deteriorate
Solution Approach 1:
The trajectory computation is segmented into fast collision-free path querying using pre-computed roadmaps and separate local optimization steps. This segmentation allows the system to quickly obtain feasible paths without exhaustive optimization, then apply localized optimization only where necessary, achieving real-time responsiveness while maintaining adequate trajectory quality for safe and efficient operation.
Data Source
AI summary
Techniques are disclosed for a trajectory planning of robots, such as collaborative robots (cobots). A controller of a robot may include a path planner, a trajectory generator, and a trajectory controller. The path planner may determine a plurality of waypoints defining a path between an initial pose of the robot and a goal pose of the robot. The trajectory generator may determine a trajectory between the initial pose and the goal pose based on the waypoints and one or more trajectory criterion. The trajectory controller may generate a control signal to control the robot based on the determined trajectory.