To develop escape manoeuvre from obstacles and to find new waypoints for dynamic trajectory modification of UAV, a novel Particle Swarm Optimization based Collision Avoidance algorithm (PSO-CA) is presented in this paper. The proposed “obstacle sense and avoid” algorithm and the logical decision-making system aids the Unmanned Aerial Vehicle (UAV) to re-route its current path to a safer flight course when an obstacle pops up along its intended path. A radar with 10 km range identifies obstacles and the UAV manoeuvres based on radar data, making it suitable for any unknown environment. The proposed system would manoeuvre the UAV autonomously along optimized alternate path to avoid the conflicting traffic. New waypoints are identified and the waypoint list is modified dynamically to avoid collision with stationary threats like hill, tree and moving intruders like other UAVs. The proposed algorithm steers the vehicle safely along alternate path with less change in intended trajectory while avoiding all potential threats. As the PSO-CA algorithm detects obstacles and identifies alternate path well in advance for unknown pop up threats, the UAV is safe and is suitable for real time environment. The proposed algorithm has considered obstacles with different positions, different sizes and random motion. Experimental results conducted on 6-DOF model UAV with different obstacles clearly justify the efficiency of the proposed method in comparison with other planners.