Abstract:This paper addresses human-robot collaboration (HRC) challenges of integrating predictions of human activity to provide a proactive-n-reactive response capability for the robot. Prior works that consider current or predicted human poses as static obstacles are too nearsighted or too conservative in planning, potentially causing delayed robot paths. Alternatively, time-varying prediction of human poses would enable robot paths that avoid anticipated human poses, synchronized dynamically in time and space. Herein, a proactive path planning method, denoted STAP, is presented that uses spatiotemporal human occupancy maps to find robot trajectories that anticipate human movements, allowing robot passage without stopping. In addition, STAP anticipates delays from robot speed restrictions required by ISO/TS 15066 speed and separation monitoring (SSM). STAP also proposes a sampling-based planning algorithm based on RRT* to solve the spatio-temporal motion planning problem and find paths of minimum expected duration. Experimental results show STAP generates paths of shorter duration and greater average robot-human separation distance throughout tasks. Additionally, STAP more accurately estimates robot trajectory durations in HRC, which are useful in arriving at proactive-n-reactive robot sequencing.
What problem does this paper attempt to address?
This paper attempts to solve the problem of robot path planning in human - robot collaboration (HRC), especially how to effectively predict human activities and react synchronously in time to avoid conflicts with human activities. Specifically, the paper proposes a method called Spatio - Temporal Avoidance of Predictions (STAP), which aims to generate robot paths by predicting human actions, thereby reducing the interference between robots and humans during task execution.
### Main problems:
1. **Myopic path planning**: Existing methods usually regard human postures as static obstacles, which leads to the myopia of path planning. That is, the robot may get into a temporary predicament because it does not consider the dynamic changes of humans, resulting in production delays.
2. **Handling of dynamic obstacles**: Existing methods fail to fully consider the time - varying nature of dynamic obstacles, resulting in path planning that is not flexible enough to effectively avoid potential future collisions.
3. **Compliance with the ISO/TS 15066 standard**: In human - robot collaboration, robots need to comply with the Speed and Separation Monitoring (SSM) rules of the ISO/TS 15066 standard. These rules require robots to slow down or stop when approaching humans to ensure safety. Existing path planning methods fail to effectively predict the impact of these speed limits on path time.
### Solutions:
1. **Spatio - temporal occupancy map**: The STAP method first generates a spatio - temporal occupancy map, which encodes the predicted human postures and their occupancy times. In this way, the robot can predict the position of humans at future time points, thereby avoiding potential collisions.
2. **RRT* - based path planning algorithm**: STAP proposes a sampling path planning algorithm based on RRT*, which can consider the spatio - temporal occupancy map and SSM rules to find the path with the minimum expected duration.
3. **Safety - aware cost function**: The cost function of STAP takes into account the SSM rules in the ISO/TS 15066 standard and predicts the impact of human postures on robot speed, thereby estimating the execution time of the path more accurately.
### Experimental results:
The experimental results show that the paths generated using the STAP method have an average execution time that is 47% less than that of other methods, and in human - robot collaboration tasks, the average human - robot separation distance of the robot is greater, improving safety. In addition, the path time estimates generated by the STAP method are also more accurate, with a smaller deviation from the actual execution time.
In conclusion, this paper solves the problems of myopia in path planning and handling of dynamic obstacles in human - robot collaboration by proposing the STAP method, while ensuring the safety and efficiency of robot paths.