Trajectplan

Trajectplan is a term used to describe the process of creating a trajectplan, a time-parameterized plan for an agent to move from a starting state to a goal state within a defined environment. Commonly used in robotics, autonomous transportation, computer animation, and logistics, a trajectplan combines a geometric route with a temporal profile, specifying where the agent should be at each moment and how fast it should move. The concept is often called trajectory planning in English, with trajectplan appearing in some European languages as a cognate.

A trajectplan must satisfy constraints such as obstacle avoidance, kinematic and dynamic limits, actuation capabilities, and timing requirements. It may aim to minimize an objective such as travel time, energy, or jerk. Plans can be open-loop, remaining fixed, or closed-loop, adapting in real time through feedback.

Common techniques include sampling-based methods such as Rapidly-exploring Random Trees (RRT) and Probabilistic Roadmaps (PRM); optimization-based approaches like CHOMP, TrajOpt, and STOMP; and graph-search methods such as A with time expansion. In practice, trajectplanning often uses model predictive control or smoothing to produce executable trajectories.

Applications span mobile robots, robotic manipulators, drones, autonomous vehicles, character animation, and logistics. Key challenges include computational complexity, uncertainty in sensing, and coordinating multiple agents in dynamic environments; ongoing work focuses on real-time performance and robust planning under disturbance.