CHAPTERS

LAUNCH VEHICLE GUIDANCE, PRESENT SCENARIO

AND FUTURE TRENDS
5.1

INTRODUCTION
Primary function of any guidance scheme is to steer the vehicle along an

optimal path, satisfying the constraints on the trajectory. A variety of guidance

schemes are adopted worldwide to cater to different launch vehicle missions. This
chapter summarizes the different guidance strategies used for steering command
generation in launch vehicles, starting from simple open loop guidance schemes to
complex adaptive guidance schemes capable of on-line trajectory reshaping,
retargeting or reconfiguration of guidance law to meet the complex performance and
mission requirements, typical of Reusable Launch Vehicle (RLV) missions. It is
very important that the selected guidance algorithm meets required end conditions
and path constraints within the specified guidance error budget of the mission.
Launch vehicles considered here include conventional Expendable Launch
Vehicles (ELVs) and advanced launch vehicles like Reusable Launch Vehicles
(RLVs) and Air-Breathing hypersonic Launch Vehicles (ABLVs). These vehicles
have typical flight paths depending upon their mission objectives. While a rocketpowered vehicle leaves the dense atmosphere quickly to minimize the drag losses,
an ABLV dwells much longer in the dense atmosphere for collecting oxygen from
the atmosphere for air-breathing propulsion. An RLV has to perform orbitaVsub-

67

orbital flight and then fly back and land on a selected site. A variety of guidance
schemes are adopted worldwide to cater to different launch vehicle missions.
Role of guidance in launch vehicle missions is to generate steering
commands for guiding the vehicle along an optimal path satisfying both path
constraints and end constraints on the trajectory. In the case of an ELV, guidance
generates steering commands from lift off till payload injection and also during post
injection manures to avoid collision of spent stage with the spacecraft after
separation. In the case of an RLV, guidance has to generate steering commands
from lift off till touch down, ie till the vehicle lands safely on a selected land site.
Optimal path as applied to launch vehicles means, fuel optimal path during powered
flight as the case is with satellite launch vehicles. For an un-powered flight, as the
case with aerodynamically controlled flight phases of an RLV, the optimal path is
any path that minimizes a suitable performance index for a selected mission. For
example, the performance indices for the descent phase of an aerodynamically
controlled RLV can be, (i) minimizing the velocity at touch down, (ii) minimizing
the heat flux,(iii)minimizing the dynamic pressure, (iv)minimizing the structural
load factor or (v) minimizing the dispersions from trim boundary on angle of attack.
Path constraints can be any vehicle or mission related constraints. While the end
constraints are in the form of desired orbital conditions for an ELV, desired touch
down conditions form the end constraints for an RLV that has to land safely on a
selected land site. It is very important that the selected guidance algorithm meets

68

required end conditions and path constraints within the specified error budget of
guidance for a mission.
This chapter addresses the current practice in the design and development of
guidance schemes for ELVs, briefly touching upon the design inputs, desirable
features, classification and selection for a typical launch vehicle mission. Challenges
involved in the guidance law development for advanced launch vehicles like RLV
and ABLY are also highlighted in this chapter. This chapter in fact summarizes the
different guidance philosophies used for steering command generation for aerospace
vehicles, starting from simple open loop guidance schemes to complex adaptive
guidance schemes capable of on-line trajectory reshaping, retargeting or
reconfiguration of guidance law to meet the complex performance and mission
requirement, common to RLV missions.

5.2

EXPENDABLE

LAUNCH

VEHICLE

GUIDANCE,

PRESENT

SCENARIO
Current practice in Expendable Lunch Vehicle (ELV) guidance technique is
addressed here. ISRO has had a number of successful Expendable Launch Vehicle
(ELV) missions to Sun Synchronous Polar Orbits (SSPOs) used for remote sensing
applications and Geo-synchronous Transfer Orbits (GTOs) used for communication
applications. The robustness and accuracy of the guidance scheme plays a major
role in injecting the satellites precisely into the required orbits for accomplishing
successful missions even in the presence of off nominal vehicle performances. This

69

section is devoted to throw some light on the various aspects involved in the choice
and design of guidance schemes for a particular launch vehicle mission.
5.2.1 Launch Vehicle Guidance
Guidance technique involves, knowing present position, direction and
magnitude of motion with respect to other points of reference, the determination of
changes in motion required to achieve a desired result. Guidance technology as
applied to launch vehicle involves, generating steering commands from lift off to
payload injection and also during post injection maneuvers.
5.2.2 Functions of Guidance
Primary function of a guidance scheme is to steer the vehicle along optimal
path, satisfying the constraints on trajectory. Guidance has to ensure that the desired
orbital/target conditions are met with minimum errors, through least effort.
Guidance also has to generate thrust cut off on meeting the required target
conditions.
5.2.3 Design Inputs for Guidance
Basic inputs for guidance design are
(i) Mission Requirements,
(ii) Constraints to be met and

(iii) Guidance Specifications.

Block diagram of guidance design inputs is shown in fig.5.l.

70

Mission
Requirements

___.lL

, ..

Guidance
Algorithm
Design

Constraints to
be met
._____ _ _

Guidance
Specifications

Fig.5.1 Guidance Design Inputs

(i)Mission Requirements: Typical requirement can be, Injecting payloads into a
Sun Synchronous Polar Orbit (SSPO), Geo-synchronous Transfer Orbit (GTO),
Medium Earth Orbit (MEO) or Low Earth Orbit (LEO).
(ii)Guidance Specifications: Typically, Specifications on orbital parameters such
as Apogee altitude, Perigee Altitude, Argument of perigee, Longitude of ascending
node, and inclination are to be met.
(iii)Constraints to be met: Include constraints related to trajectory, mission and
vehicle structural integrity.
Trajectory constraints: Eg.Spent Stage impact constraints.
Mission constraints: Eg.Maximum heat flux, maximum dynamic pressure.
Vehicle Constraints: Eg.Maximum thermal load, aerodynamic load, steering
command rates.

71

5.2.4 Desirable Features of Guidance

Some of the desirable features of guidance are
(i) Robustness to large dispersions
(ii) Flexibility to handle different missions (LEO to GTO) or (SSPO to GTO)
(iii) Fault detection and isolation

(iv) Mission Salvage plan to save the mission.

5.2.5 Guidance Design Methodology
Guidance design starts once the mission requirements, constraints and
guidance specifications are clearly defined. Guidance schemes can be broadly
classified into two, Open Loop Guidance (OLG) and Closed Loop Guidance (CLG)
as shown in fig.5.2.

Guidance

CLG

OLG

Exo-Atmospheric phase

Atmospheric phase

Fig.5.2 Guidance Design Methodology

In OLG, an optimal steering program is computed in ground (predetermined) from an accurate model of the vehicle system and stored on-board.
Constraints on path, loads on the vehicle (dynamic pressure & angle of attack) and.

72

heating constraints are taken into account in ground-based design. Steering

commands are stored on-board as a look up table and generated as function of
current time or altitude.
In CLG, steering commands are computed on-board, in real-time making use
of Navigation information on (i)current states of the vehicle (position & velocity)
and (ii) the desired target states stored on-board. This requires the optimal solution
of equations of motion governing the path of a Launch Vehicle. The equations of
motion are nonlinear and time varying. All the end conditions, either initial or final,
are not completely defined. This results in nonlinear time varying Two Point
Boundary Value Problem (TPBVP) that is to be solved for determining the steering
commands. For obtaining the solutions in real-time using less powerful on-board
computers simplifying assumptions are used for solutions. Typical assumptions are
(i) Constant thrust, (ii) Constant mass flow rate and (iii)Small angle approximations.
Assumptions (i) & (ii) are valid for liquid and cryogenic stages ,but not true for
solid stages.
Selection of OLG or CLG depends on mission requirements. If the allowed
orbital dispersions in terms of altitude, velocity and inclination are high and the
vehicle dispersions are Iowan OLG can be employed. If the orbital dispersion
allowed is stringent, CLG need be used.
Current practice is to clear the atmosphere using OLG, an optimal open loop
trajectory designed in ground, for minimum load on the vehicle conditions.
Generate steering commands in CLG mode in exo-atmospheric phase using fuel

73

optimal schemes. Any deviations in states achieved at the end of atmospheric phase
will be corrected by upper stage guidance schemes in CLG mode.

5.2.6 Classification of Closed Loop Guidance Schemes for Exo-Atmospheric

phase
Closed Loop Guidance Schemes for Launch Vehicles can be broadly
classified as (i)Explicit /Path adaptive schemes and(ii) Implicit/Perturbation
schemes.

(i) Explicit schemes

Explicit schemes are also called Path adaptive schemes, since they obtain the
solution of equations of motion repeatedly at points along the resulting trajectory
and the current and target states appear explicitly in the solution. Explicit schemes
are very fuel optimal and accurate but are characterized by large on-board
computational needs. Some examples of this guidance algorithms are, Explicit-E
Guidance(E), Flat Earth Guidance (FE).

(ii) Implicit schemes

Implicit schemes are also called perturbation schemes. They assume that the
mission is completely defined before launch and a reference nominal trajectory is
available. The problem is to find the optimum steering logic that forces the vehicle
to closely follow the nominal trajectory. These schemes can achieve good accuracy
at low on-board computational needs backed up by large pre-computations. Some
examples of this guidance schemes are, Velocity to be gained (Vg) algorithm,
Linear perturbation guidance, Delta Guidance, Q guidance.

74

5.2.7 Guidance Schemes Present Scenario in Launch Vehicles

Current trend in launch vehicle missions is to use a judicious mix of explicit,
implicit and open loop schemes. Guidance Schemes developed and flight proven in
launch vehicles for SSPO and GTO missions, include E-Guidance, FE-Guidance,
Vg-guidance. E and FE are explicit schemes while Vg is an implicit guidance
scheme. E and FE schemes are best suited for liquid engines, where as V g is most
suitable for solid motors, for generating the steering commands. Basic features of
the above mentioned guidance algorithms are brought out here.
(i)

Explicit E-Guidance
E-Guidance algorithm (George. W. Cherry.,1964)works on required

acceleration concept. The steering law is based on complete solution of equations of

motion (spherical Earth model used) repeatedly along the flight path. This algorithm
is the most fuel optimal scheme but complexity is more.
E-Guidance problem obtains split solution for thrust allocation along
(i)radial,(ii),horizontal and (iii) perpendicular directions.

Solution is arrived at

through an iteration for time- to- go(TGO) and this parameter ensures that required
states in all three directions are met simultaneously. Block diagram of E-Guidance
is shown in fig.5.3

75

R,V,G

RD,VD

,,
,
RADIAL CO- ORDINATE
CONTROL
TGO

TIME TOGO
ESTIMATOR
VOo

--

VOt=

,,

HORIWNTAL
SPEED CONTROL

TGO

FigS.3 E-Goidance Block Diagram

TOO-- time- to- go
VBo- Target horizontal velocity
VO..- Predicted horizontal velocity
R,V,G- Current radial distance, velocity and acceleration due to gravity
RD.VD- Target radial distance and velocity
OR- steering command
E-Guidance is used as final stage guidance for achieving precise orbital
injection for SSPO missions. This scheme is also used as second stage guidance for
GTO missions for accurately controlling the spent stage impact.

76

(ii)

Flat Earth( FE) Guidance

This is a path adaptive scheme (M. J. W., Gage,1968) that uses an

approximate solution to equations of motion, for simplicity Flat Earth (uniform

gravitational field) is assumed, for generating the steering commands. Though this
scheme is not as fuel optimal as E-Guidance, computationally it is very simple. FE
scheme also works on required acceleration concept. The simplicity is achieved by
avoiding iteration for TGO and by Flat Earth assumption. This makes the scheme
slightly sub-optimal, which is overcome by applying appropriate correction factors
to the steering commands. Block diagram of FE-Guidance is shown in fig.S.4

R,V,G

TOO

RD,VD

RADIAL COORDINATE
CONTROL

HORIZONTAL
SPEED CONTROL

90
Fig.S.4 FE-Guidance Block Diagram
TOO- time- to- go
V9n- Target horizontal velocity

R. V,G- Current radial distance, velocity and acceleration due to gravity

RD,VD- Target radial distance and velocity

77

FE Guidance is used in Cryogenic final stage for achieving the target orbit
conditions of GTO, which is an ellipse with fixed orientation in inertial space.
(iii) Velocity to be gained( V g) Guidance
Vg guidance (George. W. Cherry.,1964) algorithm works on the required
velocity concept. Steering commands are generated using cross product steering
law. This scheme is very simple but gives fairly accurate results with appropriate
design of steering command gains. Using this algorithm it is possible to insert the
vehicle in an elliptical orbit with a fixed angular momentum, but it is not possible to
control the states of point of injection. The block schematic is shown fig.5.5.

HD,RD

l
-

COMPUTE
VR

+--

COMPUTE
VG=VR-V

9c 'lfC

roc=K(VGxAT)

Fig.5.5 Vg-Guidance Block Diagram

HD,RD- angular momentum and apogee radial distance of target ellipse
Vo- Velocity to be gained
VR- Required velocity
V- Current velocity from navigation
AT- Thrust acceleration

78

K - Steering command Gain

O>e- Vehicle Body rates

Be

,"'e -Pitch & yaw steering commands.

Vg Guidance algorithm is used for generating the steering commands in

solid propelled third stage for SSPO mission.

5.3

ADVANCED LAUNCH VEHICLE GUIDANCE, FUTURE TRENDS

New challenges lie in the design of guidance schemes for advanced launch

vehicle missions like, Reusable Launch Vehicle (RLV) missions and Air-Breathing
Launch Vehicle (ABLV) mission. This is because the guidance requirements are
more complex for these kinds of missions. The closed loop guidance schemes we
discussed so far for ELVs in previous sections are all Exo-atmospheric phase/
vacuum phase guidance techniques. We were using only simple open loop guidance
schemes for steering the ELVs within the atmosphere. Unlike ELVs, RLV and
ABLV dwell in dense atmosphere for longer periods to meet their mission
objectives. This calls for atmospheric phase closed loop guidance technique for
ensuring the atmospheric constraints on dynamic pressure, heat flux and load factor
and. perform controlled flight, effectively meeting the required end conditions.
Following section gives an overview of guidance schemes used for reentry
mission ofRLV and air-breathing ascent hypersonic flight mission of ABLV.
5.3.1 Re-entry Guidance for RLV
Flight profile of a typical Reusable Launch Vehicle mission is different from
that of an Expendable Launch Vehicle mission. The ascent phase is similar to ELV

79

ascent phase mission and all the guidance schemes discussed in earlier sections are
applicable. During descent phase, the RLV reenters the atmosphere at very high
velocities. This requires that appropriate energy management be performed for
reducing the velocity to reasonable levels for landing.
Another important fact is that the vehicle has to be guided during descent
phase to fly within an entry corridor bounded by constraints on dynamic pressure,
heat flux, normal load factor and equilibrium glide (to ensure controllability). This
is achieved by the re-entry guidance scheme that guides RLV to fly within the entry
corridor and reach the specified landing site for safe landing with correct level of
energy.
The entry guidance problem concerns with providing steering commands to
control the entry trajectory from the atmospheric penetration to a target point where
the vehicle has sufficient energy for approach and landing. The Shuttle entry
guidance (H.arpold, J.

c.,

and Gavert, D.E.,1983), concept is to design a reference

drag acceleration profile as a piecewise analytical function of earth relative velocity.

The downrange distance can then be analytically predicted, and the reference drag
profile is adjusted as necessary for updated ranging along the trajectory. A linear,
time varying, feedback control law is used to track the reference drag profile.
Ronneke.A. J., and Markl.A (1994) propose to schedule the drag profile as functions
of the energy for the entire trajectory for more accurate range prediction. Ping Lu
. (1997), proposes to design the reference drag profile as a function of the energy. A

80

nonlinear feedback trajectory control law based on a predictive control method is

derived by Ping.Lu, (1994 ) for trajectory tracking of the reference drag profile.
Maintaining all strengths of the Shuttle entry guidance scheme, it would
appear most logical that the entry guidance for typical RLV would adopt concept of
commanded drag based algorithms. At the same time, incorporation of results of
recent research and the increased capabilities of avionics into the RLV guidance
scheme may further improve the performance, provide greater flexibility, and
increase the control capability to accommodate contingencies and larger operating
domain. The block diagram of entry guidance applicable for RLV type mission is
shown in fig.5.6.
Store Drer, <lrer
as function of
Vr

Predict downrange

Dest, adjust drag profile to

achieve desired downran2e

Compute Aa to
follow Des1

<Xe= <lrer+ Aa

Fig.5.6. RLV Entry Guidance

Drat - Reference drag as a function of relative velocity
Vr - relative velocity
<Xret - nominal angle of attack
Dest - Eestimated drag

81

<Xe - Commanded angle of attack

Reference drag profile is designed as a function of relative velocity and
stored on-board. Nominal angle of attack profile is also stored on board. The
constraints on dynamic pressure (q), qa, heat flux(Q) and load factor are mapped on
to the Drag vs relative velocity plane. The down range is predicted in each guidance
cycle. The drag profile is adjusted so that the desired down range is achieved. Now
change in angle of attack (L\a) required to follow the commanded drag is computed
and the angle of attack command is updated based on that.

5.3.2 Adaptive Guidance for ABLV

Air-Breathing Launch Vehicles (ABLVs) are being considered as promising
candidates for the future low cost space transportation systems. These vehicles
provide more effective way to launch large spacecrafts to Low Earth Orbit (LEO)
than Expendable Rocket Launch Vehicles (ERLVs). An ABLV is an advanced
launch vehicle that reaches hypersonic velocities by using a new form of propulsion
called air-breathing propulsion. The payload capability of these vehicles is one order
in magnitude greater than that of ERLVs. These vehicles can take off and land like
an aircraft and at the same time reach high velocities (hypersonic velocities,
Mach>7). The most challenging aspect of guidance for ABLV is that, the thrust
developed by an air-breathing engine depends on flight path. In fact the thrust
depends on Mach number (ie.,h-altitude,V-velocity) and angle of attack a. Hence
new techniques of guidance are required for guiding the vehicle during air-breathing
ascent phase of ABLV.

82

The near minimum fuel trajectory for an ABLV is, however, substantially
different from that of a rocket powered expendable launch vehicle. Whereas a
rocket-powered vehicle leaves the dense atmosphere quickly to minimize the drag
losses, an ABLV dwells much longer in the dense atmosphere where the air
breathing propulsion is more efficient. The amount of uncertainties in aerodynamic
parameters is significantly greater for ABLVs compared to expendable rocket
launch vehicles. Due to above reasons guidance and control technology dependent
on pre-launch, predetermined trajectory as used in conventional launch vehicles is
inadequate for ABLVs.
The most adaptive approach to guidance problem requires no nominal
trajectory and in fact continuously selects a new path for the remainder of the flight
based only on the current state vector, the constraints to be met and the desired end
conditions. There is only one part to the guidance algorithm and that is computed
on-board with the feedback from the navigation system. Using this approach the
guidance is not constrained to follow a trajectory computed prior to flight. Instead it
can take advantage of dispersions that aid the flight and minimize the dispersions
that detract from the flight. The on-board computational requirements to perform
this function are more for this approach. But with present day flight computers these
difficulties are disappearing.
Another technique used in adaptive guidance is the development of on-line
trajectory reshaping algorithms. This approach uses multi guidance loops for

83

generating the steering commands. The architecture of Adaptive Guidance strategy

used for ABLV is shown in fig.5.7.

rt

On-line
Trajectory
Generation
R*,V*,T*

R,V,a
1

Adaptive
Guidance
Law

cl>c

Propulsion

Vehicle

...
"

Control
Power Plant

<7:c

,,.
R,V,a
Navigation
Fig.5.7. Architecture of Adaptive Guidance Scheme for ABLV
R,V- Current position and velocity,
R*,V*- Optimal trajectory
T*- Optimal thrust history
a - Current Angle of attack
cl>c- Throttle control
U- Lift control input
T c - Thrust command
a c -Commanded angle of attack

84

The optimal trajectory re-shaping algorithm updates the trajectory at longer

intervals ( 1 Os) of time. This is the outer guidance loop. The adaptive guidance
commands are generated by the inner loop guidance at shorter intervals of
time(O. l s). The inner loop guidance module includes a gain adaptation algorithm
that varies the feedback gains on altitude and altitude rate errors, depending on the
current estimate of errors from Navigation. Use of a Proportional plus Derivative
(PD) controller to track the optimal trajectory generated on-line, has been found to
give good robust and stable guidance performance. The performance of the
controller depends upon the values of gains, Kp-position gain and Kci-velocity gain.
These gains are estimated on-line by the guidance algorithm as functions of altitude
and altitude rate errors in actual flight. The guidance law controls the vehicle lift
using the PD controller. The design and development and validation of adaptive
guidance scheme for a typical ABLY is described in chapter 8.

5.4

SUMMARY
This chapter presented the current practice and the future trends of launch

vehicle guidance design. Current practice in design and development of guidance

algorithms for Expendable Launch Vehicles (ELVs) was addressed, briefly
highlighting the design inputs, desirable features, classification and selection for a
particular mission. Challenges involved in the guidance law synthesis for advanced
launch vehicles like Reusable Launch Vehicle (RLV) and Air-Breathing Launch
Vehicle (ABLY) are also discussed. This chapter has summarized the different
guidance philosophies that are used for steering command generation for launch

85

vehicles, starting from simple open loop guidance schemes to complex adaptive
guidance schemes capable of on-line trajectory reshaping to meet the mission
requirements ofRLV and ABLV.
We have seen that the explicit guidance techniques require analytical
solutions for onboard implementation. It is possible to get analytical solution by the
application of simplifying assumptions that are true for flight in vacuum phase.
Typical RLVs perform flight within atmosphere for longer durations, and that
includes both atmospheric ascent as well as descent. Present practice of open loop
guidance scheme within atmospheric phase is inadequate for this class of vehicles.
From re usability and reliability point of view, RLVs require closed loop guidance
algorithms to steer the vehicle within atmosphere. To take care of uncertainties in
vehicle & atmospheric parameters, these schemes have to be adaptive and robust to
such dispersions. Simplicity and flexibility to handle different missions and phases
of flight using a single integrated guidance scheme is one of the most desirable
feature of RLV guidance. This has motivated the development of a new Integrated
Adaptive Guidance algorithm from lift off to touch down of a typical RLV mission.

86