These are most probably for an article which is released at the said alt be it 30 km or 40 km by a booster.
As for RudraMII or RudraMIII, its terminal vel is subsonic at max range as borne out by the specs of warheads.
RudraMIII has a max range of 550 km but at this range its terminal speed is under 300 m/s. If it’s used for a range of 400 km, then it’s terminal speed is just above M 1.
This optimisation model is for a specific phase of flight for operation within a specific range bracket and the model allows the user to customise trajectory shaping as per the mission choice as well as on demand basis via the user interface input.
Typically the atmospheric boost glide vehicle operates at high speed but mostly in the high supersonic regime and differ slightly from the HBGV. So when the choice of propulsion is a single solid rocket motor with a limited time of burn, this option becomes very promising due to the shorter range bracket envelope.
Compared to a fully powered flight at a certain altitude, optimising for glide mode has the advantage of not being time constrained since the solid motor can burn for a specific duration only. The propellant composition, burn rate etc can be for short or long duration, but the time for each choice is finite ie the motor will work for x or y sec duration. So to extract better flight performance like range & provide the unpredictability of flight path from the same flight vehicle, optimisation of trajectory is essential.
About the warhead of the rudrams, not at all like that. The fuzing system used are impact fuzes but there is slight ms level delay involved otherwise mission objective will not be met. Since warhead options are different, pcb fcb pf various types, missile roles are different in ARM and ground attack too. So mission plan is also done accordingly and it is associated with angle of attack.
For ARM role , even against fixed radar site or an AD zone where mobile AD system (multiple TELs) is deployed in scattered way the AOA is shallow but not too much else the miss distance becomes unacceptable, CEP must be < 25-30m. Hence an AOA is chosen 40-45deg which keeps the miss distance within acceptable tolerance and the pre frag warhead can do the desired damage even if CEP at 30m. The missile can glide at supersonic speed till target and impact still >1mach. But try making the missile fly at 30deg or below angle of attack approach to target, you will see more drag and it will not be able to have desired miss distance, CEP will be . 60 , 70m even more due to closing in to target at very low angle. No course correction possible in those last few sec.
In dive from top mode, final approach only minor course correction, and impact speed 550m/s near mach 2 at a steep 80-85deg angle. Here the role is like a sledgehammer, missile will penetrate and fuze activates within few mili sec resulting in blast after penetration. warhead is also shaped accordingly. PF like a fat jug and pcb like a bullet. In case of rudram 3 one warhead size small as missile speed does the penetration job aka high terminal velocity likely 550-600m/s with slight post impact delay fuzing that survives thru impact then blast to take out inside.
Quite sure terminal speed will vary in same way for different trajectory shapes ie prioritizing different primary constraint factor each time. This is actually where the development stage work happens, those algos can have many flaws and create condition due to data fed via sensor that degrade the performance. Hence the dev team would make necessary fixes, change parameter in cases. Much like software version rollout.