Trajectory Planning - GGP
Trajectory planning ·The trajectory planning in joint space and Cartesian space is completed under the condition of specifying some motion parameters of the robot, such as speed, acceleration, motion time, etc. Manual selection of speed or time cannot guarantee the motion performance of the robot. The speed can be either too high or low, and that is not conducive to the performance of the robot. In multi-point planning, although the problem of cumbersome setting of intermediate point speed has been solved, for the motion with many intermediate points, the setting of time in each time period is also a large amount of work.
By adjusting the parameters of PSO, particle swarm algorithm and combining the random search characteristics of GA, genetic algorithm, an improved genetic particle swarm optimization algorithm, GGP for robot fast trajectory planning is proposed.
For the planning of multi-point joint trajectory, four indicators are selectedthe optimization model: motion time, energy consumption, velocity limiting and acceleration limiting. If the weighting factors are $\omega_1$, $\omega_2$, $\omega_3$, $\omega_4$ respectively, then the designed fitness function is:
The particle velocity and position formula of the improved PSO algorithm is updated as follows:
\[\begin{array}{c} v_{i j}(k+1)=\omega(k) v_{i j}(k)+c_{1}(k) R_{1}\left(p_{i j}(k)-x_{i j}(k)\right)+c_{2}(k) R_{2}\left(p_{g j}(k)-x_{i j}(k)\right) \\ \\ x_{i j}(k+1)=x_{i j}(k)+v_{i j}(k+1) \end{array}\]where $\omega(k)$ stands for an adaptive adjustment of weight, $c_{1}(k)$ and $c_{2}(k)$ is the updated learning factor equation.
The GGP algorithm is applied to automatic trajectory planning of robots, giving full play to its high-dimensional constraints, fast and simple characteristics. To verify the effectiveness and advantages of this algorithm, the convergence characteristics and test time of this algorithm are analyzed and compared below.
