Collision Avoidance from Multiple Passive Agents with Partially Predictable Behavior

Navigating a robot in a dynamic environment is a challenging task. Most modern applications require navigation of robots among humans, vehicles and other robots. Robots working in such dynamic environments must be able to successfully navigate to goal positions while multiple agents are moving around them. In most of the applications of navigating an agent, the moving obstacle{s} has free will, and their future behavior is only partially predictable (if at all). When facing such situation, obstacle's path must be predicted using on line prediction techniques, so that robot is not blinded to future collisions. When~the on-line path prediction is used to plan motion, it is likely that the model of the future that is obtained will have limited time duration. Therefore, a planner is required that takes into account the validity duration of the model of the environment and allowable time for computing a collision free path. Also, it is necessary to consider various sources of uncertainties present, e.g., passive agents unpredictability, uncertainty in resulting state under a given control action, and~localization error. Further, almost all of the mobile robots in the real world application{s} are subjected to kinodynamic constraints. Considering the kinodynamic constraints of robot while planning adds an extra challenge for the problem of navigating a robot in dynamic environment. In this study, we have explored a navigational strategy for collision avoidance of a kinodynamically constrained robot from multiple moving passive agents with partially predictable behavior. Specifically, this work presents a new approach to identify the set of control inputs to the robot, named control obstacle, which leads it towards a collision with a passive agent moving along an arbitrary path. The proposed method is developed by generalizing the concept of nonlinear velocity obstacle (NLVO), which is used to avoid collision with a passive agent, and takes into account the kinodynamic constraints on robot motion. Further, this work formulates the navigational problem as an optimization problem, which allows the robot to make a safe decision in the presence of various sources of unmodelled uncertainties. In this study, the performance of the proposed algorithm is evaluated for different parameters and is compared to existing velocity obstacle-based approaches. The simulated experiments show the excellent performance of the proposed approach in term of computation time and success rate. As this work proposes a local navigation approach, it requires a global motion planner to converge to goal in the presence of the large static obstacles. This work study the one possible implementation of the global navigational plan which results in model predictive partial motion planning framework.Docto