Continuous-time Gaussian Process Trajectory Generation for Multi-robot Formation via Probabilistic Inference

Shuang Guo,Bo Liu,Shen Zhang,Jifeng Guo,Changhong Wang

DOI: https://doi.org/10.48550/arXiv.2010.13148

2021-07-30

Abstract:In this paper, we extend a famous motion planning approach GPMP2 to multi-robot cases, yielding a novel centralized trajectory generation method for the multi-robot formation. A sparse Gaussian Process model is employed to represent the continuous-time trajectories of all robots as a limited number of states, which improves computational efficiency due to the sparsity. We add constraints to guarantee collision avoidance between individuals as well as formation maintenance, then all constraints and kinematics are formulated on a factor graph. By introducing a global planner, our proposed method can generate trajectories efficiently for a team of robots which have to get through a width-varying area by adaptive formation change. Finally, we provide the implementation of an incremental replanning algorithm to demonstrate the online operation potential of our proposed framework. The experiments in simulation and real world illustrate the feasibility, efficiency and scalability of our approach.

Robotics

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to generate continuous - time trajectories in multi - robot systems. In particular, when a multi - robot formation passes through areas with varying widths, how to efficiently calculate goal - oriented, collision - free trajectories that satisfy kinematic and formation constraints. Specifically, the paper proposes a novel centralized trajectory generation method based on the sparse Gaussian process model for multi - robot formations. This method represents the trajectory optimization problem on a factor graph and uses probabilistic inference to solve it quickly. At the same time, a global planner is introduced to adaptively change the formation to deal with passages of different widths. In addition, the paper also implements an incremental replanning algorithm, demonstrating the potential of the proposed framework for online operation.

### Main Contributions
1. **Extension of GPMP2**: The paper extends the well - known motion planning method GPMP2 to the multi - robot scenario and proposes a new centralized multi - robot formation trajectory generation method.
2. **Sparse Gaussian Process Model**: The continuous - time trajectories of all robots are represented using the sparse Gaussian process model, which improves computational efficiency.
3. **Global Planner**: A global planner, including formation planning and task allocation, is introduced to adaptively calculate a suitable rectangular formation and its execution time intervals.
4. **Incremental Replanning Algorithm**: An incremental replanning algorithm is implemented, demonstrating the potential of the proposed framework for online operation.

### Technical Details
- **Formation Planning**: The rectangular safe flight corridor (RSFC) is used to generate the expected formation, and the execution time intervals of each formation are allocated according to the length of the path segments.
- **Task Allocation**: A task allocation algorithm for rectangular formations is proposed to reduce conflicts during formation transitions and ensure that the optimizer can find feasible trajectories more quickly.
- **Trajectory Optimization**: The trajectory optimization problem is represented on a factor graph, with constraints such as kinematics, obstacle avoidance, inter - individual collision avoidance, and formation maintenance added, and is solved using nonlinear least - squares optimization.
- **Incremental Replanning**: Based on iGPMP2, the current solution is updated incrementally to achieve fast trajectory updates and support online operations.

### Experimental Verification
The paper verifies the effectiveness and feasibility of the proposed method through simulation and practical experiments, demonstrating the efficiency and robustness of a multi - rotor UAV team when passing through areas with varying widths. The experimental results show that this method can quickly calculate the complete trajectories of 10 robots in complex tasks, with good computational efficiency and scalability.

### Conclusion
The paper proposes a novel multi - robot formation trajectory generation framework. Through the combination of global planning and trajectory optimization, the problem of trajectory generation when multi - robot systems pass through areas with varying widths is solved. The experimental results verify the effectiveness and feasibility of this method and demonstrate its potential in practical applications.